Jiajun Wu 0001

EDBT 2026 Demo / reviewers in the wild / expert

dblp:117/4768 · DBLP profile ↗

← Back

214ranked-venue papers

15as first author

144since 2021 · last 2026

0000-0002-4176-343XORCID · conflict

Domains — the database's venue-derived domains; a paper can count in several

Artificial intelligence and machine learning · 199 · 15 first-author · 132 since 2021Graphics, computer vision, multimedia, augmented reality and games · 102 · 9 first-author · 79 since 2021Systems, architecture and hardware · 17 · 10 since 2021Applied, interdisciplinary, general and emerging computing · 8 · 1 first-author · 3 since 2021Human-computer interaction and ubiquitous computing · 2 · 1 since 2021

Year	Publication	Venue	Position
2026	ZeroHSI: Zero-Shot 4D Human-Scene Interaction by Video Generation abstract Human-scene interaction (HSI) generation is crucial for applications in embodied AI, virtual reality, and robotics. Yet, existing methods cannot synthesize interactions in unseen environments such as in-the-wild scenes or reconstructed scenes, as they rely on paired 3D scenes and captured human motion data for training, which are unavailable for unseen environments. We present ZeroHSI, a novel approach that enables zero-shot 4D human-scene interaction synthesis, eliminating the need for training on any MoCap data. Our key insight is to distill human-scene interactions from state-of-the-art video generation models, which have been trained on vast amounts of natural human movements and interactions,and use differentiable rendering to reconstruct human-scene interactions. ZeroHSI can synthesize realistic human motions in both static scenes and environments with dynamic objects, without requiring any ground-truth motion data. We evaluate ZeroHSI on a curated dataset of different types of various indoor and outdoor scenes with different interaction prompts, demonstrating its ability to generate diverse and contextually appropriate human-scene interactions. Project page: https://awfuact.github.io/zerohsi. Hong-Xing Yu, Jiaman Li, Jiajun Wu 0001	3DV	4
2026	3D-GENERALIST: Vision-Language-Action Models for Crafting 3D Worlds abstract Creating 3D graphics content for immersive and interactive worlds remains labor-intensive, limiting our ability to create large-scale synthetic data for training foundation models. Recent methods aim to alleviate this, but they often focus on a single aspect (e.g., layout) and do not improve generation quality by simply scaling computational resources. We recast 3D environment generation as a sequential decision-making problem, using Vision-Language Models (VLMs) as policies that output actions to jointly craft a 3D environment's layout, materials, lighting, and assets. Our framework, 3D-Generalist, trains VLMs to generate more prompt-aligned 3D environments via self-improvement fine-tuning. We demonstrate the effectiveness of 3D-Generalist and our training strategy in generating simulation-ready 3D environments. We also demonstrate its quality and scalability for synthetic data generation by pretraining a vision foundation model on the generated data. After fine-tuning on downstream tasks, we show that it surpasses models pre-trained on meticulously human-crafted synthetic data and approaches results achieved when training with orders of magnitude larger real data. Fan-Yun Sun, Shengguang Wu, Christian Jacobsen, Thomas Yim, Haoming Zou, Alexander Zook, Shangru Li, Yu-Hsin Chou, Ethem Can, Xunlei Wu, Clemens Eppner, Valts Blukis, Jonathan Tremblay, Jiajun Wu 0001, Stanley T. Birchfield, Nick Haber	3DV	14
2026	10 Open Challenges Steering the Future of Vision-Language-Action Models abstract Due to their ability of follow natural language instructions, vision-language-action (VLA) models are increasingly preva- lent in the embodied AI arena, following the widespread suc- cess of their precursors—LLMs and VLMs. In this paper, we discuss 10 principal milestones in the ongoing develop- ment of VLA models—multimodality, reasoning, data, eval- uation, cross-robkot action generalization, efficiency, whole- body coordination, safety, agents, and coordination with hu- mans. Furthermore, we discuss the emerging trends of us- ing spatial understanding, modeling world dynamics, post training, and data synthesis—all aiming to reach these mile- stones. Through these discussions, we hope to bring attention to the research avenues that may accelerate the development of VLA models into wider acceptability. Soujanya Poria, Navonil Majumder, Chia-Yu Hung, Amir Ali Bagherzadeh, Kenneth Kwok, Ziwei Wang 0010, Cheston Tan, Jiajun Wu 0001, David Hsu	AAAI	9
2026	Discovering Hybrid World Representations with Co-Evolving Foundation Models abstract This perspective article discusses an emerging research direction: to what extent can foundation models yield usable structure for modeling the physical world? We offer a Markovian formulation of structured world models and outline the notion of multi-level hybrid world representations that support compositional structure. We then review and suggest possible discovery paradigms, spanning distillation, interaction-driven continual learning, and ensemble learning. Jiajun Wu 0001, Hong-Xing Yu, Joy Hsu, Jiayuan Mao	AAAI	1
2025	Birth and Death of a Rose abstract We study the problem of generating temporal object intrinsics—temporally evolving sequences of object geometry, reflectance, and texture, such as a blooming rose—from pre-trained 2D foundation models. Unlike conventional 3D modeling and animation techniques that require extensive manual effort and expertise, we introduce a method that generates such assets with signals distilled from pre-trained 2D diffusion models. To ensure the temporal consistency of object intrinsics, we propose Neural Templates for temporal-state-guided distillation, derived automatically from image features from self-supervised learning. Our method can generate high-quality temporal object intrinsics for several natural phenomena and enable the sampling and controllable rendering of these dynamic objects from any viewpoint, under any environmental lighting conditions, at any time of their lifespan. Project website: https://chen-geng.com/rose4d. Chen Geng 0001, Shangzhe Wu, Jiajun Wu 0001	CVPR	4
2025	Diffusion Self-Distillation for Zero-Shot Customized Image Generation abstract Text-to-image diffusion models produce impressive results but are frustrating tools for artists who desire fine-grained control. For example, a common use case is to create images of a specific concept in novel contexts, i.e., "identity-preserving generation". This setting, along with many other tasks (e.g., relighting), is a natural fit for image+text-conditional generative models. However, there is insufficient high-quality paired data to train such a model directly. We propose Diffusion Self-Distillation, a method for using a pre-trained text-to-image model to generate its own dataset for text-conditioned image-to-image tasks. We first leverage a text-to-image diffusion model’s in-context generation ability to create grids of images and curate a large paired dataset with the help of a vision-language model. We then fine-tune the text-to-image model into a text+image-to-image model using the curated paired dataset. We demonstrate that Diffusion Self-Distillation outperforms existing zero-shot methods and is competitive with per-instance tuning techniques on a wide range of identity-preserving generation tasks, without requiring test-time optimization. Project page: primecai.github.io/dsd. Shengqu Cai, Eric R. Chan, Leonidas J. Guibas, Jiajun Wu 0001, Gordon Wetzstein	CVPR	5
2025	Digital Twin Catalog: A Large-Scale Photorealistic 3D Object Digital Twin Dataset abstract We introduce Digital Twin Catalog (DTC), a new large-scale photorealistic 3D object digital twin dataset. A digital twin of a 3D object is a highly detailed, virtually indistinguishable representation of a physical object, accurately capturing its shape, appearance, physical properties, and other attributes. Recent advances in neural-based 3D reconstruction and inverse rendering have significantly improved the quality of 3D object reconstruction. Despite these advancements, there remains a lack of a large-scale, digital twin quality real-world dataset and benchmark that can quantitatively assess and compare the performance of different reconstruction methods, as well as improve reconstruction quality through training or fine-tuning. Moreover, to democratize 3D digital twin creation, it is essential to integrate creation techniques with next-generation egocentric computing platforms, such as AR glasses. Currently, there is no dataset available to evaluate 3D object reconstruction using egocentric captured images. To address these gaps, the DTC dataset features 2,000 scanned digital twin-quality 3D objects, along with image sequences captured under different lighting conditions using DSLR cameras and egocentric AR glasses. This dataset establishes the first comprehensive real-world evaluation benchmark for 3D digital twin creation tasks, offering a robust foundation for comparing and improving existing reconstruction methods. The DTC dataset is already released at https://www.projectaria.com/datasets/dtc/ and we will also make the baseline evaluations open-source. Zhao Dong 0001, Ka Chen, Zhaoyang Lv, Hong-Xing Yu, Yufeng Zhu, Stephen Tian, Zhengqin Li, Geordie Moffatt, Sean Christofferson, James Fort, Xiaqing Pan, Mingfei Yan, Jiajun Wu 0001, Carl Yuheng Ren, Richard A. Newcombe	CVPR	15
2025	FluidNexus: 3D Fluid Reconstruction and Prediction from a Single Video abstract We study reconstructing and predicting 3D fluid appearance and velocity from a single video. Current methods require multi-view videos for fluid reconstruction. We present FluidNexus, a novel framework that bridges video generation and physics simulation to tackle this task. Our key insight is to synthesize multiple novel-view videos as references for reconstruction. FluidNexus consists of two key components: (1) a novel-view video synthesizer that combines frame-wise view synthesis with video diffusion refinement for generating realistic videos, and (2) a physics-integrated particle representation coupling differentiable simulation and rendering to simultaneously facilitate 3D fluid reconstruction and prediction. To evaluate our approach, we collect two new real-world fluid datasets featuring textured backgrounds and object interactions. Our method enables dynamic novel view synthesis, future prediction, and interaction simulation from a single fluid video. Project website: https://yuegao.me/FluidNexus. Hong-Xing Yu, Bo Zhu 0002, Jiajun Wu 0001	CVPR	4
2025	PGC: Physics-Based Gaussian Cloth from a Single Pose abstract We introduce a novel approach to reconstruct simulation-ready garments with intricate appearance. Despite recent advancements, existing methods often struggle to balance the need for accurate garment reconstruction with the ability to generalize to new poses and body shapes or require large amounts of data to achieve this. In contrast, our method only requires a multi-view capture of a single static frame. We represent garments as hybrid mesh-embedded 3D Gaussian splats, where the Gaussians capture near-field shading and high-frequency details, while the mesh encodes far-field albedo and optimized reflectance parameters. We achieve novel pose generalization by exploiting the mesh from our hybrid approach, enabling physics-based simulation and surface rendering techniques, while also capturing fine details with Gaussians that accurately reconstruct garment details. Our optimized garments can be used for simulating garments on novel poses, and garment relighting. Project page: phys-gaussian-cloth.github.io. Michelle Guo, Matt Jen-Yuan Chiang, Igor Santesteban, Nikolaos Sarafianos, Hsiao-Yu Chen, Oshri Halimi, Aljaz Bozic, Shunsuke Saito, Jiajun Wu 0001, C. Karen Liu, Tuur Stuyck, Egor Larionov	CVPR	9
2025	Category-Agnostic Neural Object Rigging abstract The motion of deformable 4D objects lies in a low-dimensional manifold. To better capture the low dimensionality and enable better controllability, traditional methods have devised several heuristic-based methods, i.e., rigging, for manipulating dynamic objects in an intuitive fashion. However, such representations are not scalable due to the need for expert knowledge of specific categories. Instead, we study the automatic exploration of such low-dimensional structures in a purely data-driven manner. Specifically, we design a novel representation that encodes deformable 4D objects into a sparse set of spatially grounded blobs and an instance-aware feature volume to disentangle the pose and instance information of the 3D shape. With such a representation, we can manipulate the pose of 3D objects intuitively by modifying the parameters of the blobs, while preserving rich instance-specific information. We evaluate the proposed method on a variety of object categories and demonstrate the effectiveness of the proposed framework. Project page: https://guangzhaohe.com/canor. Guangzhao He, Chen Geng 0001, Shangzhe Wu, Jiajun Wu 0001	CVPR	4
2025	Lifting Motion to the 3D World via 2D Diffusion abstract Estimating 3D motion from 2D observations is a longstanding research challenge. Prior work typically requires training on datasets containing ground truth 3D motions, limiting their applicability to activities well-represented in existing motion capture data. This dependency particularly hinders generalization to out-of-distribution scenarios or subjects where collecting 3D ground truth is challenging, such as complex athletic movements or animal motion. We introduce MVLift, a novel approach to predict global 3D motion—including both joint rotations and root trajectories in the world coordinate system—using only 2D pose sequences for training. Our multi-stage framework leverages 2D motion diffusion models to progressively generate consistent 2D pose sequences across multiple views, a key step in recovering accurate global 3D motion. MVLift generalizes across various domains, including human poses, human-object interactions, and animal poses. Despite not requiring 3D supervision, it outperforms prior work on five datasets, including those methods that require 3D supervision. Jiaman Li, C. Karen Liu, Jiajun Wu 0001	CVPR	3
2025	LayoutVLM: Differentiable Optimization of 3D Layout via Vision-Language Models abstract Spatial reasoning is a fundamental aspect of human cognition, enabling intuitive understanding and manipulation of objects in three-dimensional space. While foundation models demonstrate remarkable performance on some benchmarks, they still struggle with 3D reasoning tasks like arranging objects in space according to open-ended language instructions, particularly in dense and physically constrained environments. We introduce LayoutVLM, a framework and scene layout representation that exploits the semantic knowledge of Vision-Language Models (VLMs) and supports differentiable optimization to ensure physical plausibility. LayoutVLM employs VLMs to generate two mutually reinforcing representations from visually marked images, and a self-consistent decoding process to improve VLMs spatial planning. Our experiments show that LayoutVLM addresses the limitations of existing LLM and constraint-based approaches, producing physically plausible 3D layouts better aligned with the semantic intent of input language instructions. We also demonstrate that fine-tuning VLMs with the proposed scene layout representation extracted from existing scene datasets can improve their reasoning performance. Fan-Yun Sun, Siyi Gu, Dylan Lim, Goutam Bhat, Federico Tombari, Manling Li, Nick Haber, Jiajun Wu 0001	CVPR	9
2025	Re-thinking Temporal Search for Long-Form Video Understanding abstract Efficiently understanding long-form videos remains a significant challenge in computer vision. In this work, we revisit temporal search paradigms for long-form video understanding and address a fundamental issue pertaining to all state-of-the-art (SOTA) long-context vision-language models (VLMs). Our contributions are twofold: First, we frame temporal search as a Long Video Haystack problem – finding a minimal set of relevant frames (e.g., one to five) from tens of thousands based on specific queries. Upon this formulation, we introduce LV-Haystack, the first dataset with 480 hours of videos, 15,092 human-annotated instances for both training and evaluation aiming to improve temporal search quality and efficiency. Results on LV-HAYSTACK highlight a significant research gap in temporal search capabilities, with current SOTA search methods only achieving 2.1% temporal F1score on the LongVideoBench subset.Next, inspired by visual search in images, we propose a lightweight temporal search framework, T* that reframes costly temporal search as spatial search. T* leverages powerful visual localization techniques commonly used in images and introduces an adaptive zooming-in mechanism that operates across both temporal and spatial dimensions. Extensive experiments show that integrating T* with existing methods significantly improves SOTA long-form video understanding. Under an inference budget of 32 frames, T* improves GPT-4o’s performance from 50.5% to 53.1% and LLaVAOneVision-OV-72B’s performance from 56.5% to 62.4% on the LongVideoBench XL subset. Our code, benchmark, and models are provided in the Supplementary material. Jinhui Ye, Zihan Wang 0008, Haosen Sun, Keshigeyan Chandrasegaran, Zane Durante, Cristóbal Eyzaguirre, Yonatan Bisk, Juan Carlos Niebles, Ehsan Adeli-Mosabbeb, Li Fei-Fei 0001, Jiajun Wu 0001, Manling Li	CVPR	11
2025	WonderWorld: Interactive 3D Scene Generation from a Single Image abstract We present WonderWorld, a novel framework for interactive 3D scene generation that enables users to interactively specify scene contents and layout and see the created scenes in low latency. The major challenge lies in achieving fast generation of 3D scenes. Existing scene generation approaches fall short of speed as they often require (1) progressively generating many views and depth maps, and (2) time-consuming optimization of the scene geometry representations. Our approach does not need to generate multiple views, and it leverages a geometry-based initialization that significantly reduces optimization time. Another challenge is generating coherent geometry that allows all scenes to be connected. We introduce the guided depth diffusion that allows partial conditioning of depth estimation. WonderWorld creates connected and diverse 3D scenes, each generated in less than 10 seconds on a single A6000 GPU, enabling real-time user interaction and exploration. We release full code, software, and interactive demos in https://kovenyu.com/WonderWorld/. Hong-Xing Yu, Haoyi Duan, Charles Herrmann, William T. Freeman, Jiajun Wu 0001	CVPR	5
2025	The Scene Language: Representing Scenes with Programs, Words, and Embeddings abstract We introduce the Scene Language, a visual scene representation that concisely and precisely describes the structure, semantics, and identity of visual scenes. It represents a scene with three key components: a ${\color{Orange}program}$ that specifies the hierarchical and relational structure of entities in the scene, ${\color{Blue}words}$ in natural language that summarize the semantic class of each entity, and ${\color{Purple}embeddings}$ that capture the visual identity of each entity. This representation can be inferred from pre-trained language models via a training-free inference technique, given text or image inputs. The resulting scene can be rendered into images using traditional, neural, or hybrid graphics renderers. Together, this forms an automated system for high-quality 3D and 4D scene generation. Compared with existing representations like scene graphs, our proposed Scene Language generates complex scenes with higher fidelity, while explicitly modeling the scene structures to enable precise control and editing. Zizhang Li, Matt Zhou, Shangzhe Wu, Jiajun Wu 0001	CVPR	5
2025	X-Capture: An Open-Source Portable Device for Multi-Sensory Learning abstract Understanding objects through multiple sensory modalities is fundamental to human perception, enabling cross-sensory integration and richer comprehension. For AI and robotic systems to replicate this ability, access to diverse, high-quality multi-sensory data is critical. Existing datasets are often limited by their focus on controlled environments, simulated objects, or restricted modality pairings. We introduce X-Capture, an open-source, portable, and cost-effective device for real-world multi-sensory data collection, capable of capturing correlated RGBD images, tactile readings, and impact audio. With a build cost under $1,000, X-Capture democratizes the creation of multi-sensory datasets, requiring only consumer-grade tools for assembly. Using X-Capture, we curate a sample dataset of 3,000 total points on 500 everyday objects from diverse, real-world environments, offering both richness and variety. Our experiments demonstrate the value of both the quantity and the sensory breadth of our data for both pretraining and fine-tuning multi-modal representations for object-centric tasks such as cross-sensory retrieval and reconstruction. X-Capture lays the groundwork for advancing human-like sensory representations in AI, emphasizing scalability, accessibility, and real-world applicability. Samuel Clarke, Suzannah Wistreich, Yanjie Ze, Jiajun Wu 0001	ICCV	4
2025	WorldScore: A Unified Evaluation Benchmark for World Generation abstract We introduce the WorldScore benchmark, the first unified benchmark for world generation. We decompose world generation into a sequence of next-scene generation tasks with explicit camera trajectory-based layout specifications, enabling unified evaluation of diverse approaches from 3D and 4D scene generation to video generation models. The WorldScore benchmark encompasses a curated dataset of 3,000 test examples that span diverse worlds: static and dynamic, indoor and outdoor, photorealistic and stylized. The WorldScore metrics evaluate generated worlds through three key aspects: controllability, quality, and dynamics. Through extensive evaluation of 19 representative models, including both open-source and closed-source ones, we reveal key insights and challenges for each category of models. Our dataset, evaluation code, and leaderboard can be found at https://haoyi-duan.github.io/WorldScore/ Haoyi Duan, Hong-Xing Yu, Li Fei-Fei 0001, Jiajun Wu 0001	ICCV	5
2025	Wonderplay: Dynamic 3D Scene Generation From a Single Image and Actions Zizhang Li, Hong-Xing Yu, Yin Yang 0002, Charles Herrmann, Gordon Wetzstein, Jiajun Wu 0001	ICCV	7
2025	Flow to the Mode: Mode-Seeking Diffusion Autoencoders for State-of-the-Art Image Tokenization abstract Since the advent of popular visual generation frameworks like VQGAN and latent diffusion models, state-of-the-art image generation systems have generally been two-stage systems that first tokenize or compress visual data into a lower-dimensional latent space before learning a generative model. Tokenizer training typically follows a standard recipe in which images are compressed and reconstructed subject to a combination of MSE, perceptual, and adversarial losses. Diffusion autoencoders have been proposed in prior work as a way to learn end-to-end perceptually-oriented image compression, but have not yet shown state-of-the-art performance on the competitive task of ImageNet-1K reconstruction. We propose FlowMo, a transformer-based diffusion autoencoder that achieves a new state-of-the-art for image tokenization at multiple compression rates without using convolutions, adversarial losses, spatially-aligned two-dimensional latent codes, or distilling from other tokenizers. Our key insight is that FlowMo training should be broken into a mode-matching pre-training stage and a mode-seeking post-training stage. In addition, we conduct extensive analyses and explore the training of generative models atop the FlowMo tokenizer. Our code and models will be available at http://kylesargent.github.io/flowmo . Kyle Sargent, Kyle Hsu, Justin Johnson 0001, Li Fei-Fei 0001, Jiajun Wu 0001	ICCV	5
2025	Weakly-Supervised Learning of Dense Functional Correspondences Stefan Stojanov, Linan Zhao, Dan Yamins, Jiajun Wu 0001	ICCV	5
2025	Range, not Independence, Drives Modularity in Biologically Inspired Representations abstract Why do biological and artificial neurons sometimes modularise, each encoding a single meaningful variable, and sometimes entangle their representation of many variables? In this work, we develop a theory of when biologically inspired networks---those that are nonnegative and energy efficient---modularise their representation of source variables (sources). We derive necessary and sufficient conditions on a sample of sources that determine whether the neurons in an optimal biologically-inspired linear autoencoder modularise. Our theory applies to any dataset, extending far beyond the case of statistical independence studied in previous work. Rather we show that sources modularise if their support is ``sufficiently spread''. From this theory, we extract and validate predictions in a variety of empirical studies on how data distribution affects modularisation in nonlinear feedforward and recurrent neural networks trained on supervised and unsupervised tasks. Furthermore, we apply these ideas to neuroscience data, showing that range independence can be used to understand the mixing or modularising of spatial and reward information in entorhinal recordings in seemingly conflicting experiments. Further, we use these results to suggest alternate origins of mixed-selectivity, beyond the predominant theory of flexible nonlinear classification. In sum, our theory prescribes precise conditions on when neural activities modularise, providing tools for inducing and elucidating modular representations in brains and machines. William Dorrell, Kyle Hsu, Luke Hollingsworth, Jin Hwa Lee, Jiajun Wu 0001, Chelsea Finn, Peter E. Latham, Timothy Edward John Behrens, James C. R. Whittington	ICLR	5
2025	What Makes a Maze Look Like a Maze? abstract A unique aspect of human visual understanding is the ability to flexibly interpret abstract concepts: acquiring lifted rules explaining what they symbolize, grounding them across familiar and unfamiliar contexts, and making predictions or reasoning about them. While off-the-shelf vision-language models excel at making literal interpretations of images (e.g., recognizing object categories such as tree branches), they still struggle to make sense of such visual abstractions (e.g., how an arrangement of tree branches may form the walls of a maze). To address this challenge, we introduce Deep Schema Grounding (DSG), a framework that leverages explicit structured representations of visual abstractions for grounding and reasoning. At the core of DSG are schemas—dependency graph descriptions of abstract concepts that decompose them into more primitive-level symbols. DSG uses large language models to extract schemas, then hierarchically grounds concrete to abstract components of the schema onto images with vision-language models. The grounded schema is used to augment visual abstraction understanding. We systematically evaluate DSG and different methods in reasoning on our new Visual Abstractions Benchmark, which consists of diverse, real-world images of abstract concepts and corresponding question-answer pairs labeled by humans. We show that DSG significantly improves the abstract visual reasoning performance of vision-language models, and is a step toward human-aligned understanding of visual abstractions. Joy Hsu, Jiayuan Mao, Josh Tenenbaum, Noah D. Goodman, Jiajun Wu 0001	ICLR	5
2025	Predicate Hierarchies Improve Few-Shot State Classification abstract State classification of objects and their relations is core to many long-horizon tasks, particularly in robot planning and manipulation. However, the combinatorial explosion of possible object-predicate combinations, coupled with the need to adapt to novel real-world environments, makes it a desideratum for state classification models to generalize to novel queries with few examples. To this end, we propose PHIER, which leverages predicate hierarchies to generalize effectively in few-shot scenarios. PHIER uses an object-centric scene encoder, self-supervised losses that infer semantic relations between predicates, and a hyperbolic distance metric that captures hierarchical structure; it learns a structured latent space of image-predicate pairs that guides reasoning over state classification queries. We evaluate PHIER in the CALVIN and BEHAVIOR robotic environments and show that PHIER significantly outperforms existing methods in few-shot, out-of-distribution state classification, and demonstrates strong zero- and few-shot generalization from simulated to real-world tasks. Our results demonstrate that leveraging predicate hierarchies improves performance on state classification tasks with limited data. Emily Jin, Joy Hsu, Jiajun Wu 0001	ICLR	3
2025	Why Is Spatial Reasoning Hard for VLMs? An Attention Mechanism Perspective on Focus Areas abstract Large Vision Language Models (VLMs) have long struggled with spatial reasoning tasks. Surprisingly, even simple spatial reasoning tasks, such as recognizing “under” or “behind” relationships between only two objects, pose significant challenges for current VLMs. We believe it is crucial to use the lens of mechanism interpretability, opening up the model and diving into model’s internal states to examine the interactions between image and text tokens during spatial reasoning. Our analysis of attention behaviors reveals significant differences in how VLMs allocate attention to image versus text. By tracing the areas of images that receive the highest attention scores throughout intermediate layers, we observe a notable pattern: errors often coincide with attention being misdirected towards irrelevant objects within the image. Moreover, such attention patterns exhibit substantial differences between familiar (e.g., “on the left side of ”) and unfamiliar (e.g.,“in front of ”) spatial relationships. Motivated by these findings, we propose ADAPTVIS based on inference-time confidence scores to sharpen the attention on highly relevant regions when the model exhibits high confidence, while smoothing and broadening the attention window to consider a wider context when confidence is lower. This training-free decoding method shows significant improvement (e.g., up to a 50 absolute point improvement) on spatial reasoning benchmarks such as WhatsUp and VSR with negligible additional cost. Shiqi Chen 0002, Tongyao Zhu, Ruochen Zhou, Jinghan Zhang 0006, Siyang Gao, Juan Carlos Niebles, Mor Geva, Junxian He, Jiajun Wu 0001, Manling Li	ICML	9
2025	Understanding Complexity in VideoQA via Visual Program Generation abstract We propose a data-driven approach to analyzing query complexity in Video Question Answering (VideoQA). Previous efforts in benchmark design have relied on human expertise to design challenging questions, yet we experimentally show that humans struggle to predict which questions are difficult for machine learning models. Our automatic approach leverages recent advances in code generation for visual question answering, using the complexity of generated code as a proxy for question difficulty. We demonstrate that this measure correlates significantly better with model performance than human estimates. To operationalize this insight, we propose an algorithm for estimating question complexity from code. It identifies fine-grained primitives that correlate with the hardest questions for any given set of models, making it easy to scale to new approaches in the future. Finally, to further illustrate the utility of our method, we extend it to automatically generate complex questions, constructing a new benchmark that is 1.9 times harder than the popular NExT-QA. Cristóbal Eyzaguirre, Igor Vasiljevic, Achal Dave, Jiajun Wu 0001, Rares Ambrus, Thomas Kollar, Juan Carlos Niebles, Pavel Tokmakov	ICML	4
2025	UAD: Unsupervised Affordance Distillation for Generalization in Robotic Manipulation abstract Understanding fine-grained object affordances is imperative for robots to manipulate objects in unstructured environments given open-ended task instructions. However, existing methods of visual affordance predictions often rely on manually annotated data or conditions only on a predefined set of tasks. We introduce Unsupervised Affordance Distillation (UAD), a method for distilling affordance knowledge from foundation models into a task-conditioned affordance model without any manual annotations. By leveraging the complementary strengths of large vision models and vision-language models, UAD automatically annotates a large-scale dataset with detailedpairs. Training only a lightweight task-conditioned decoder atop frozen features, UAD exhibits notable generalization to in-the-wild robotic scenes and to various human activities, despite only being trained on rendered objects in simulation. Using affordance provided by UAD as the observation space, we show an imitation learning policy that demonstrates promising generalization to unseen object instances, object categories, and even variations in task instructions after training on as few as 10 demonstrations. Project website with Appendix: unsup-affordance.github.io/. Yihe Tang, Wenlong Huang, Yingke Wang, Chengshu Li 0001, Roy Yuan, Jiajun Wu 0001, Li Fei-Fei 0001	ICRA	7
2025	Learning Smooth Humanoid Locomotion through Lipschitz-Constrained Policies abstract Reinforcement learning combined with sim-to-real transfer offers a general framework for developing locomotion controllers for legged robots. To facilitate successful deployment in the real world, smoothing techniques, such as low-pass filters and smoothness rewards, are often employed to develop policies with smooth behaviors. However, because these techniques are non-differentiable and usually require tedious tuning of a large set of hyperparameters, they tend to require extensive manual tuning for each robotic platform. To address this challenge and establish a general technique for enforcing smooth behaviors, we propose a simple and effective method that imposes a Lipschitz constraint on a learned policy, which we refer to as Lipschitz-Constrained Policies (LCP). We show that the Lipschitz constraint can be implemented in the form of a gradient penalty, which provides a differentiable objective that can be easily incorporated with automatic differentiation frameworks. We demonstrate that LCP effectively replaces the need for smoothing rewards or low-pass filters and can be easily integrated into training frameworks for many distinct humanoid robots. We extensively evaluate LCP in both simulation and real-world humanoid robots, producing smooth and robust locomotion controllers. All simulation and deployment code, along with complete checkpoints, is available on our project page: https://lipschitz-constrained-policy.github.io. Xialin He, Yen-Jen Wang, Qiayuan Liao, Yanjie Ze, Zhongyu Li 0003, S. Shankar Sastry, Jiajun Wu 0001, Koushil Sreenath, Xue Bin Peng	IROS	8
2025	Generalizable Humanoid Manipulation with 3D Diffusion Policies abstract Humanoid robots capable of autonomous operation in diverse environments have long been a goal for roboticists. However, autonomous manipulation by humanoid robots has largely been restricted to one specific scene, primarily due to the difficulty of acquiring generalizable skills and the expensiveness of in-the-wild humanoid robot data. In this work, we build a real-world robotic system to address this challenging problem. Our system is mainly an integration of 1) a whole-upper-body robotic teleoperation system to acquire human-like robot data, 2) a 25-DoF humanoid robot platform with a height-adjustable cart and a 3D LiDAR sensor, and 3) an improved 3D Diffusion Policy learning algorithm for humanoid robots to learn from noisy human data. We run more than 2000 episodes of policy rollouts on the real robot for rigorous policy evaluation. Empowered by this system, we show that using only data collected in one single scene and with only onboard computing, a full-sized humanoid robot can autonomously perform skills in diverse real-world scenarios. Videos are available at humanoid-manipulation.github.io. Yanjie Ze, Xialin He, Xue Bin Peng, Jiajun Wu 0001	IROS	8
2025	From Programs to Poses: Factored Real-World Scene Generation via Learned Program Libraries abstract Real-world scenes, such as those in ScanNet, are difficult to capture, with highly limited data available. Generating realistic scenes with varied object poses remains an open and challenging task. In this work, we propose FactoredScenes, a framework that synthesizes realistic 3D scenes by leveraging the underlying structure of rooms while learning the variation of object poses from lived-in scenes. We introduce a factored representation that decomposes scenes into hierarchically organized concepts of room programs and object poses. To encode structure, FactoredScenes learns a library of functions capturing reusable layout patterns from which scenes are drawn, then uses large language models to generate high-level programs, regularized by the learned library. To represent scene variations, FactoredScenes learns a program-conditioned model to hierarchically predict object poses, and retrieves and places 3D objects in a scene. We show that FactoredScenes generates realistic, real-world rooms that are difficult to distinguish from real ScanNet scenes. Joy Hsu, Emily Jin, Jiajun Wu 0001, Niloy J. Mitra	NeurIPS	3
2025	Taming generative video models for zero-shot optical flow extraction abstract Extracting optical flow from videos remains a core computer vision problem. Motivated by the recent success of large general-purpose models, we ask whether frozen self-supervised video models trained only to predict future frames can be prompted, without fine-tuning, to output flow. Prior attempts to read out depth or illumination from video generators required fine-tuning; that strategy is ill-suited for flow, where labeled data is scarce and synthetic datasets suffer from a sim-to-real gap. Inspired by the Counterfactual World Model (CWM) paradigm, which can obtain point-wise correspondences by injecting a small tracer perturbation into a next-frame predictor and tracking its propagation, we extend this idea to generative video models for zero-shot flow extraction. We explore several popular architectures and find that successful zero-shot flow extraction in this manner is aided by three model properties: (1) distributional prediction of future frames (avoiding blurry or noisy outputs); (2) factorized latents that treat each spatio-temporal patch independently; and (3) random-access decoding that can condition on any subset of future pixels. These properties are uniquely present in the recently introduced Local Random Access Sequence (LRAS) architecture. Building on LRAS, we propose KL-tracing: a novel test-time inference procedure that injects a localized perturbation into the first frame, rolls out the model one step, and computes the Kullback–Leibler divergence between perturbed and unperturbed predictive distributions. Without any flow-specific fine-tuning, our method is competitive with state-of-the-art, task-specific models on the real-world TAP-Vid DAVIS benchmark and the synthetic TAP-Vid Kubric. Our results show that counterfactual prompting of controllable generative video models is an effective alternative to supervised or photometric-loss methods for high-quality flow. Seungwoo Kim, Khai Loong Aw, Klemen Kotar, Cristóbal Eyzaguirre, Wanhee Lee, Yunong Liu, Jared Watrous, Stefan Stojanov, Juan Carlos Niebles, Jiajun Wu 0001, Dan Yamins	NeurIPS	10
2025	Self-Supervised Learning of Motion Concepts by Optimizing Counterfactuals abstract Estimating motion primitives from video (e.g., optical flow and occlusion) is a critically important computer vision problem with many downstream applications, including controllable video generation and robotics. Current solutions are primarily supervised on synthetic data or require tuning of situation-specific heuristics, which inherently limits these models' capabilities in real-world contexts. A natural solution to transcend these limitations would be to deploy large-scale, self-supervised video models, which can be trained scalably on unrestricted real-world video datasets. However, despite recent progress, motion-primitive extraction from large pretrained video models remains relatively underexplored. In this work, we describe Opt-CWM, a self-supervised flow and occlusion estimation technique from a pretrained video prediction model. Opt-CWM uses ``counterfactual probes'' to extract motion information from a base video model in a zero-shot fashion. The key problem we solve is optimizing the quality of these probes, using a combination of an efficient parameterization of the space counterfactual probes, together with a novel generic sparse-prediction principle for learning the probe-generation parameters in a self-supervised fashion. Opt-CWM achieves state-of-the-art performance for motion estimation on real-world videos while requiring no labeled data. Stefan Stojanov, David Wendt, Seungwoo Kim, Rahul M. V., Kevin T. Feigelis, Klemen Kotar, Khai Loong Aw, Jiajun Wu 0001, Dan Yamins	NeurIPS	8
2025	VAGEN: Reinforcing World Model Reasoning for Multi-Turn VLM Agents abstract A major challenge in training VLM agents, compared to LLM agents, is that states shift from simple texts to complex visual observations, which introduces partial observability and demands robust world modeling. We ask: can VLM agents build internal world models through explicit visual state reasoning? In this work, we architecturally enforce and reward VLM agent’s reasoning process via reinforcement learning (RL), formulating the problem as a Partially Observable Markov Decision Process (POMDP). We demonstrate that structuring agent’s reasoning into StateEstimation (“what is the current state?”) and TransitionModeling (“what is next?”) is critical by studying five reasoning strategies. Investigating how agents should ground visual states and represent these internal beliefs, we reveal the optimal representations are task-dependent: Natural Language excels at capturing semantic relationships for general tasks, while Structured formats are essential for high-precision manipulation. These insights motivate our approach to reward shaping and credit assignment. We leverage a WorldModeling Reward to densely rewards the agent’s turn-by-turn state predictions, while our Bi-Level General Advantage Estimation (Bi-Level GAE) enables turn-aware credit assignment. Through such world model reasoning, we enable a 3B model to achieve performance of 0.82 on a set of five diverse agent tasks, nearly 3× improvement over its untrained counterpart (0.21) and surpassing proprietary reasoning models like GPT-5 (0.75), Gemini 2.5 Pro (0.67) and Claude 4.5 (0.62). All experiments are supported by our VAGEN framework, a scalable system for training and analyzing multi-turn VLM agents across diverse visual environments Kangrui Wang, Pingyue Zhang, Zihan Wang 0008, Yaning Gao, Qineng Wang, Hanyang Chen, Zhengyuan Yang, Ranjay Krishna, Jiajun Wu 0001, Li Fei-Fei 0001, Yejin Choi 0001, Manling Li	NeurIPS	12
2025	Web-Scale Collection of Video Data for 4D Animal Reconstruction abstract Computer vision for animals holds great promise for wildlife research but often depends on large-scale data, while existing collection methods rely on controlled capture setups. Recent data-driven approaches show the potential of single-view, non-invasive analysis, yet current animal video datasets are limited—offering as few as 2.4K 15-frame clips and lacking key processing for animal-centric 3D/4D tasks. We introduce an automated pipeline that mines YouTube videos and processes them into object-centric clips, along with auxiliary annotations valuable for downstream tasks like pose estimation, tracking, and 3D/4D reconstruction. Using this pipeline, we amass 30K videos (2M frames)—an order of magnitude more than prior works. To demonstrate its utility, we focus on 4D quadruped animal reconstruction task. To support this task, we present Animal4D, a benchmark of 230 manually filtered sequences with 11K frames showcasing clean, diverse animal motions. We evaluate state-of-the-art model-based and model-free methods on Animal4D, finding that 2D metrics favor the former despite unrealistic 3D shapes, while the latter yields more natural reconstructions but scores lower—revealing a gap in current evaluation. To address this, we enhance a recent model-free approach with sequence-level optimization, establishing the first 4D animal reconstruction baseline. Together, our pipeline, benchmark, and baseline aim to advance large-scale, markerless 4D animal reconstruction and related tasks from in-the-wild videos. Code and datasets are available at https://github.com/briannlongzhao/Animal-Video-Processing. Brian Nlong Zhao, Jiajun Wu 0001, Shangzhe Wu	NeurIPS	2
2025	CRAFT: Designing Creative and Functional 3D Objects abstract For designing a wide range of everyday objects, the design process should be aware of both the human body and the underlying semantics of the design specification. However, these two objectives present significant challenges to the current AI-based designing tools. In this work, we present a method to synthesize body-aware 3D objects from a base mesh given an input body geometry and either text or image as guidance. The generated objects can be simulated on virtual characters, or fabricated for real-world use. We propose to use a mesh deformation procedure that optimizes for both semantic alignment as well as contact and penetration losses. Using our method, users can generate both virtual or real-world objects from text, image, or sketch, without the need for manual artist intervention. We present both qualitative and quantitative results on various object categories, demonstrating the effectiveness of our approach. Michelle Guo, Mia Tang, Hannah Cha, C. Karen Liu, Jiajun Wu 0001	WACV	6
2024	Partial-View Object View Synthesis via Filtering Inversion abstract We propose Filtering Inversion (FINV), a learning framework and optimization process that predicts a renderable 3D object representation from one or few partial views. FINV addresses the challenge of synthesizing novel views of objects from partial observations, spanning cases where the object is not entirely in view, is partially occluded, or is only observed from similar views. To achieve this, FINV learns shape priors by training a 3D generative model. At inference, given one or more views of a novel real-world object, FINV first finds a set of latent codes for the object by inverting the generative model from multiple initial seeds. Maintaining the set of latent codes, FINV filters and resamples them after receiving each new observation, akin to particle filtering. The generator is then finetuned for each latent code on the available views in order to adapt to novel objects. We show that FINV successfully synthesizes novel views of real-world objects (e.g., chairs, tables, and cars), even if the generative prior is trained only on synthetic objects. The ability to address the sim-to-real problem allows FINV to be used for object categories without real-world datasets. FINV achieves state-of-the-art performance on multiple real-world datasets, recovers object shape and texture from partial and sparse views, is robust to occlusion, and is able to incrementally improves its representation with more observations. Fan-Yun Sun, Jonathan Tremblay, Valts Blukis, Danfei Xu, Boris Ivanovic, Péter Karkus, Stanley T. Birchfield, Dieter Fox, Yunzhu Li, Jiajun Wu 0001, Marco Pavone 0001, Nick Haber	3DV	12
2024	CityPulse: Fine-Grained Assessment of Urban Change with Street View Time Series abstract Urban transformations have profound societal impact on both individuals and communities at large. Accurately assessing these shifts is essential for understanding their underlying causes and ensuring sustainable urban planning. Traditional measurements often encounter constraints in spatial and temporal granularity, failing to capture real-time physical changes. While street view imagery, capturing the heartbeat of urban spaces in a pedestrian point of view, can add as a high-definition, up-to-date, and on-the-ground visual proxy of urban change. We curate the largest street view time series dataset to date, and propose an end-to-end change detection model to effectively capture physical alterations in the built environment at scale. We demonstrate the effectiveness of our proposed method by benchmark comparisons with previous literature and implementing it at the city-wide level. Our approach has the potential to supplement existing dataset and serve as a fine-grained and accurate assessment of urban change. Zejia Wu, Jiajun Wu 0001, Jackelyn Hwang, Ram Rajagopal	AAAI	3
2024	SkyScript: A Large and Semantically Diverse Vision-Language Dataset for Remote Sensing abstract Remote sensing imagery, despite its broad applications in helping achieve Sustainable Development Goals and tackle climate change, has not yet benefited from the recent advancements of versatile, task-agnostic vision language models (VLMs). A key reason is that the large-scale, semantically diverse image-text dataset required for developing VLMs is still absent for remote sensing images. Unlike natural images, remote sensing images and their associated text descriptions cannot be efficiently collected from the public Internet at scale. In this work, we bridge this gap by using geo-coordinates to automatically connect open, unlabeled remote sensing images with rich semantics covered in OpenStreetMap, and thus construct SkyScript, a comprehensive vision-language dataset for remote sensing images, comprising 2.6 million image-text pairs covering 29K distinct semantic tags. With continual pre-training on this dataset, we obtain a VLM that surpasses baseline models with a 6.2% average accuracy gain in zero-shot scene classification across seven benchmark datasets. It also demonstrates the ability of zero-shot transfer for fine-grained object attribute classification and cross-modal retrieval. We hope this dataset can support the advancement of VLMs for various multi-modal tasks in remote sensing, such as open-vocabulary classification, retrieval, captioning, and text-to-image synthesis. Rajanie Prabha, Jiajun Wu 0001, Ram Rajagopal	AAAI	4
2024	Whodunnit? Inferring what happened from multimodal evidence Sarah A. Wu, Erik Brockbank, Hannah Cha, Jan-Philipp Fränken, Emily Jin, Zhuoyi Huang, Jiajun Wu 0001, Tobias Gerstenberg	CogSci	9
2024	Naturally Supervised 3D Visual Grounding with Language-Regularized Concept Learners abstract 3D visual grounding is a challenging task that often requires direct and dense supervision, notably the semantic label for each object in the scene. In this paper, we instead study the naturally supervised setting that learns from only 3D scene and QA pairs, where prior works underperform. We propose the Language-Regularized Concept Learner (LARC), which uses constraints from language as regularization to significantly improve the accuracy of neurosymbolic concept learners in the naturally supervised setting. Our approach is based on two core insights: the first is that language constraints (e.g., a word's relation to another) can serve as effective regularization for structured representations in neuro-symbolic models; the second is that we can query large language models to distill such constraints from language properties. We show that LARC improves performance of prior works in naturally supervised 3D visual grounding, and demonstrates a wide range of 3D visual reasoning capabilities-from zero-shot composition, to data efficiency and transferability. Our method represents a promising step towards regularizing structured visual reasonin frameworks with language-based priors, for learning in settings without dense supervision. Joy Hsu, Jiajun Wu 0001	CVPR	4
2024	BEHAVIOR Vision Suite: Customizable Dataset Generation via Simulation abstract The systematic evaluation and understanding of computer vision models under varying conditions require large amounts of data with comprehensive and customized labels, which real-world vision datasets rarely satisfy. While current synthetic data generators offer a promising alternative, particularly for embodied AI tasks, they often fall short for computer vision tasks due to low asset and rendering quality, limited diversity, and unrealistic physical properties. We introduce the BEHAVIOR Vision Suite (BVS), a set of tools and assets to generate fully customized synthetic data for systematic evaluation of computer vision models, based on the newly developed embodied AI benchmark, BEHAVIOR-1 K. BVS supports a large number of adjustable parameters at the scene level (e.g., lighting, object placement), the object level (e.g., joint configuration, attributes such as “filled” and “folded”), and the camera level (e.g., field of view, focal length). Researchers can arbitrarily vary these parameters during data generation to perform controlled experiments. We showcase three example application scenarios: systematically evaluating the robustness of models across different continuous axes of domain shift, evaluating scene understanding models on the same set of images, and training and evaluating simulation-to-real transfer for a novel vision task: unary and binary state prediction. Project website: https://behavior-vision-suite.github.io/ Yunhao Ge, Yihe Tang, Cem Gökmen, Chengshu Li 0001, Wensi Ai, Benjamin Jose Martinez, Arman Aydin, Mona Anvari, Ayush K. Chakravarthy, Hong-Xing Yu, Josiah Wong, Sanjana Srivastava, Sharon Lee, Shengxin Zha, Laurent Itti, Yunzhu Li, Roberto Martin Martin, Miao Liu 0007, Pengchuan Zhang, Li Fei-Fei 0001, Jiajun Wu 0001	CVPR	23
2024	Learning the 3D Fauna of the Web abstract Learning 3D models of all animals in nature requires massively scaling up existing solutions. With this ultimate goal in mind, we develop 3D-Fauna, an approach that learns a pan-category deformable 3D animal model for more than 100 animal species jointly. One crucial bottle-neck of modeling animals is the limited availability of training data, which we overcome by learning our model from 2D Internet images. We show that prior approaches, which are category-specific, fail to generalize to rare species with limited training images. We address this challenge by introducing the Semantic Bank of Skinned Models (SBSM), which automatically discovers a small set of base animal shapes by combining geometric inductive priors with semantic knowledge implicitly captured by an off-the-shelf self-supervised feature extractor. To train such a model, we also contribute a new large-scale dataset of diverse animal species. At inference time, given a single image of any quadruped animal, our model reconstructs an articulated 3D mesh in a feed-forward manner in seconds. Zizhang Li, Dor Litvak, Ruining Li, Tomas Jakab, Christian Rupprecht 0001, Shangzhe Wu, Andrea Vedaldi, Jiajun Wu 0001	CVPR	9
2024	ZeroNVS: Zero-Shot 360-Degree View Synthesis from a Single Image abstract We introduce a 3D-aware diffusion model, ZeroNVS, for single-image novel view synthesis for in-the-wild scenes. While existing methods are designed for single objects with masked backgrounds, we propose new techniques to address challenges introduced by in-the-wild multi-object scenes with complex backgrounds. Specifically, we train a generative prior on a mixture of data sources that capture object-centric, indoor, and outdoor scenes. To address issues from data mixture such as depth-scale ambiguity, we propose a novel camera conditioning parameterization and normalization scheme. Further, we observe that Score Distillation Sampling (SDS) tends to truncate the distribution of complex backgrounds during distillation of 360-degree scenes, and propose “SDS anchoring” to improve the diversity of synthesized novel views. Our model sets a new state-of-the-art result in LPIPS on the DTU dataset in the zero-shot setting, even outperforming methods specifically trained on DTU. We further adapt the challenging Mip-NeRF 360 dataset as a new benchmark for single-image novel view synthesis, and demonstrate strong performance in this setting. Code and models are available at this url. Kyle Sargent, Zizhang Li, Tanmay Shah, Charles Herrmann, Hong-Xing Yu, Eric R. Chan, Dmitry Lagun, Li Fei-Fei 0001, Deqing Sun, Jiajun Wu 0001	CVPR	11
2024	Hearing Anything Anywhere abstract Recent years have seen immense progress in 3D computer vision and computer graphics, with emerging tools that can virtualize real-world 3D environments for numerous Mixed Reality (XR) applications. However, alongside immersive visual experiences, immersive auditory experiences are equally vital to our holistic perception of an environment. In this paper, we aim to reconstruct the spatial acoustic characteristics of an arbitrary environment given only a sparse set of (roughly 12) room impulse response (RIR) recordings and a planar reconstruction of the scene, a setup that is easily achievable by ordinary users. To this end, we introduce DIFFRIR, a differentiable RIR rendering framework with interpretable parametric models of salient acoustic features of the scene, including sound source directivity and surface reflectivity. This allows us to synthesize novel auditory experiences through the space with any source audio. To evaluate our method, we collect a dataset of RIR recordings and music in four diverse, real environments. We show that our model outperforms state-of-the-art baselines on rendering monaural and binaural RIRs and music at unseen locations, and learns physically interpretable parameters characterizing acoustic properties of the sound source and surfaces in the scene. Mason L. Wang, Ryosuke Sawata, Samuel Clarke, Ruohan Gao, Shangzhe Wu, Jiajun Wu 0001	CVPR	6
2024	ULIP-2: Towards Scalable Multimodal Pre-Training for 3D Understanding abstract Recent advancements in multimodal pretraining have shown promising efficacy in 3D representation learning by aligning multimodal features across 3D shapes, their 2D counterparts, and language descriptions. However, the methods used by existing frameworks to curate such multimodal data, in particular language descriptions for 3D shapes, are not scalable, and the collected language descriptions are not diverse. To address this, we introduce ULIP-2, a simple yet effective tri-modal pretraining framework that leverages large multimodal models to automatically generate holistic language descriptions for 3D shapes. It only needs 3D data as input, eliminating the need for any manual 3D annotations, and is therefore scalable to large datasets. ULIP-2 is also equipped with scaled-up backbones for better multimodal representation learning. We conduct experiments on two large-scale 3D datasets, Objaverse and ShapeNet, and augment them with tri-modal datasets of 3D point clouds, images, and language for training ULIP-2. Experiments show that ULIP-2 demonstrates substantial benefits in three downstream tasks: zero-shot 3D classification, standard 3D classification with fine-tuning, and 3D captioning (3D-to-language generation). It achieves a new SOTA of 50.6% (top-1) on Objaverse-LVIS and 84.7% (top-1) on ModelNet40 in zero-shot classification. In the ScanObjectNN benchmark for standard fine-tuning, ULIP-2 reaches an overall accuracy of 91.5% with a compact model of only 1.4 million parameters. ULIP-2 sheds light on a new paradigm for scalable multimodal 3D representation learning without human annotations and shows significant improvements over existing baselines. The code and datasets are released at https://github.com/salesforce/ULIP. Le Xue, Ning Yu 0006, Shu Zhang 0007, Artemis Panagopoulou, Junnan Li 0001, Roberto Martin Martin, Jiajun Wu 0001, Caiming Xiong, Ran Xu 0001, Juan Carlos Niebles, Silvio Savarese	CVPR	7
2024	Holodeck: Language Guided Generation of 3D Embodied AI Environments abstract 3D simulated environments play a critical role in Embodied AI, but their creation requires expertise and extensive manual effort, restricting their diversity and scope. To miti-gate this limitation, we present Holodeck, a system that generates 3D environments to match a user-supplied prompt fullyautomatedly. Holodeck can generate diverse scenes, e.g., arcades, spas, and museums, adjust the designs for styles, and can capture the semantics of complex queries such as “apartment for a researcher with a cat” and “office of a professor who is a fan of Star Wars”. Holodeck leverages a large language model (i.e., GPT-4) for common sense knowledge about what the scene might look like and uses a large collection of 3D assets from Objaverse to populate the scene with diverse objects. To address the challenge of positioning objects correctly, we prompt GPT-4 to generate spatial relational constraints between objects and then optimize the layout to satisfy those constraints. Our large-scale human evaluation shows that annotators prefer Holodeck over manually designed procedural baselines in residential scenes and that Holodeck can produce high-quality outputs for diverse scene types. We also demonstrate an exciting application of Holodeck in Embodied AI, training agents to navigate in novel scenes like music rooms and daycares without human-constructed data, which is a significant step forward in developing general-purpose embodied agents. Yue Yang 0006, Fan-Yun Sun, Luca Weihs, Eli VanderBilt, Alvaro Herrasti, Winson Han, Jiajun Wu 0001, Nick Haber, Ranjay Krishna, Lingjie Liu, Chris Callison-Burch, Mark Yatskar, Aniruddha Kembhavi	CVPR	7
2024	WonderJourney: Going from Anywhere to Everywhere abstract We introduce WonderJourney, a modular framework for perpetual 3D scene generation. Unlike prior work on view generation that focuses on a single type of scenes, we start at any user-provided location (by a text description or an image), and generate a journey through a long sequence of diverse yet coherently connected 3D scenes. We leverage an LLM to generate textual descriptions of the scenes in this journey, a text-driven point cloud generation pipeline to make a compelling and coherent sequence of 3D scenes, and a large VLM to verify the generated scenes. We show compelling, diverse visual results across various scene types and styles, forming imaginary “wonder journeys ”. Project website: https://kovenyu.com/WonderJourney/. “No, no! The adventures first, explanations take such a dreadful time.“ - Alice's Adventures in Wonderland Hong-Xing Yu, Haoyi Duan, Junhwa Hur, Kyle Sargent, Michael Rubinstein, William T. Freeman, Forrester Cole, Deqing Sun, Noah Snavely, Jiajun Wu 0001, Charles Herrmann	CVPR	10
2024	Controllable Human-Object Interaction Synthesis Jiaman Li, Alexander Clegg, Roozbeh Mottaghi, Jiajun Wu 0001, Xavier Puig, C. Karen Liu	ECCV (41)	4
2024	Ponymation: Learning Articulated 3D Animal Motions from Unlabeled Online Videos Keqiang Sun, Dor Litvak, Hongsheng Li 0001, Jiajun Wu 0001, Shangzhe Wu	ECCV (1)	5
2024	3D Congealing: 3D-Aware Image Alignment in the Wild Zizhang Li, Amit Raj, Andreas Engelhardt, Yuanzhen Li, Tingbo Hou, Jiajun Wu 0001, Varun Jampani	ECCV (1)	7
2024	PhysDreamer: Physics-Based Interaction with 3D Objects via Video Generation Hong-Xing Yu, Rundi Wu, Brandon Yushan Feng, Changxi Zheng, Noah Snavely, Jiajun Wu 0001, William T. Freeman	ECCV (2)	7
2024	Reconstruction and Simulation of Elastic Objects with Spring-Mass 3D Gaussians Licheng Zhong, Hong-Xing Yu, Jiajun Wu 0001, Yunzhu Li	ECCV (2)	3
2024	Neural Polynomial Gabor Fields for Macro Motion Analysis abstract We study macro motion analysis, where macro motion refers to the collection of all visually observable motions in a dynamic scene. Traditional filtering-based methods on motion analysis typically focus only on local and tiny motions, yet fail to represent large motions or 3D scenes. Recent dynamic neural representations can faithfully represent motions using correspondences, but they cannot be directly used for motion analysis. In this work, we propose Phase-based neural polynomial Gabor fields (Phase-PGF), which learns to represent scene dynamics with low-dimensional time-varying phases. We theoretically show that Phase-PGF has several properties suitable for macro motion analysis. In our experiments, we collect diverse 2D and 3D dynamic scenes and show that Phase-PGF enables dynamic scene analysis and editing tasks including motion loop detection, motion factorization, motion smoothing, and motion magnification. Project page: https://chen-geng.com/phasepgf Chen Geng 0001, Hong-Xing Yu, Sida Peng, Xiaowei Zhou 0001, Jiajun Wu 0001	ICLR	5
2024	Patched Denoising Diffusion Models For High-Resolution Image Synthesis abstract We propose an effective denoising diffusion model for generating high-resolution images (e.g., 1024$\times$512), trained on small-size image patches (e.g., 64$\times$64). We name our algorithm Patch-DM, in which a new feature collage strategy is designed to avoid the boundary artifact when synthesizing large-size images. Feature collage systematically crops and combines partial features of the neighboring patches to predict the features of a shifted image patch, allowing the seamless generation of the entire image due to the overlap in the patch feature space. Patch-DM produces high-quality image synthesis results on our newly collected dataset of nature images (1024$\times$512), as well as on standard benchmarks of LHQ(1024$\times$ 1024), FFHQ(1024$\times$ 1024) and on other datasets with smaller sizes (256$\times$256), including LSUN-Bedroom, LSUN-Church, and FFHQ. We compare our method with previous patch-based generation methods and achieve state-of-the-art FID scores on all six datasets. Further, Patch-DM also reduces memory complexity compared to the classic diffusion models. Project page: https://patchdm.github.io. Zheng Ding, Jiajun Wu 0001, Zhuowen Tu	ICLR	3
2024	Language-Informed Visual Concept Learning abstract Our understanding of the visual world is centered around various concept axes, characterizing different aspects of visual entities. While different concept axes can be easily specified by language, e.g., color, the exact visual nuances along each axis often exceed the limitations of linguistic articulations, e.g., a particular style of painting. In this work, our goal is to learn a language-informed visual concept representation, by simply distilling large pre-trained vision-language models. Specifically, we train a set of concept encoders to encode the information pertinent to a set of language-informed concept axes, with an objective of reproducing the input image through a pre-trained Text-to-Image (T2I) model. To encourage better disentanglement of different concept encoders, we anchor the concept embeddings to a set of text embeddings obtained from a pre-trained Visual Question Answering (VQA) model. At inference time, the model extracts concept embeddings along various axes from new test images, which can be remixed to generate images with novel compositions of visual concepts. With a lightweight test-time finetuning procedure, it can also generalize to novel concepts unseen at training. Project page at https://cs.stanford.edu/~yzzhang/projects/concept-axes. Sharon Lee, Shangzhe Wu, Jiajun Wu 0001	ICLR	4
2024	Learning Planning Abstractions from Language abstract This paper presents a framework for learning state and action abstractions in sequential decision-making domains. Our framework, planning abstraction from language (PARL), utilizes language-annotated demonstrations to automatically discover a symbolic and abstract action space and induce a latent state abstraction based on it. PARL consists of three stages: 1) recovering object-level and action concepts, 2) learning state abstractions, abstract action feasibility, and transition models, and 3) applying low-level policies for abstract actions. During inference, given the task description, PARL first makes abstract action plans using the latent transition and feasibility functions, then refines the high-level plan using low-level policies. PARL generalizes across scenarios involving novel object instances and environments, unseen concept compositions, and tasks that require longer planning horizons than settings it is trained on. Joy Hsu, Jiayuan Mao, Jiajun Wu 0001	ICLR	5
2024	Tripod: Three Complementary Inductive Biases for Disentangled Representation Learning abstract Inductive biases are crucial in disentangled representation learning for narrowing down an underspecified solution set. In this work, we consider endowing a neural network autoencoder with three select inductive biases from the literature: data compression into a grid-like latent space via quantization, collective independence amongst latents, and minimal functional influence of any latent on how other latents determine data generation. In principle, these inductive biases are deeply complementary: they most directly specify properties of the latent space, encoder, and decoder, respectively. In practice, however, naively combining existing techniques instantiating these inductive biases fails to yield significant benefits. To address this, we propose adaptations to the three techniques that simplify the learning problem, equip key regularization terms with stabilizing invariances, and quash degenerate incentives. The resulting model, Tripod, achieves state-of-the-art results on a suite of four image disentanglement benchmarks. We also verify that Tripod significantly improves upon its naive incarnation and that all three of its "legs" are necessary for best performance. Kyle Hsu, Jubayer Ibn Hamid, Kaylee Burns, Chelsea Finn, Jiajun Wu 0001	ICML	5
2024	Physically Grounded Vision-Language Models for Robotic Manipulation abstract Recent advances in vision-language models (VLMs) have led to improved performance on tasks such as visual question answering and image captioning. Consequently, these models are now well-positioned to reason about the physical world, particularly within domains such as robotic manipulation. However, current VLMs are limited in their understanding of the physical concepts (e.g., material, fragility) of common objects, which restricts their usefulness for robotic manipulation tasks that involve interaction and physical reasoning about such objects. To address this limitation, we propose PHYSOBJECTS, an object-centric dataset of 39.6K crowd-sourced and 417K automated physical concept annotations of common household objects. We demonstrate that fine-tuning a VLM on PhysObjects improves its understanding of physical object concepts, including generalization to held-out concepts, by capturing human priors of these concepts from visual appearance. We incorporate this physically grounded VLM in an interactive framework with a large language model-based robotic planner, and show improved planning performance on tasks that require reasoning about physical object concepts, compared to baselines that do not leverage physically grounded VLMs. We additionally illustrate the benefits of our physically grounded VLM on a real robot, where it improves task success rates. We release our dataset and provide further details and visualizations of our results at https://iliad.stanford.edu/pg-vlm/. Jensen Gao, Bidipta Sarkar, Fei Xia 0002, Ted Xiao, Jiajun Wu 0001, Brian Ichter, Anirudha Majumdar, Dorsa Sadigh	ICRA	5
2024	Learning to Design 3D Printable Adaptations on Everyday Objects for Robot Manipulation abstract Advancements in robot learning for object manipulation have shown promising results, yet certain everyday objects remain challenging for robots to effectively interact with. This discrepancy arises from the fact that human-designed objects are optimized for human use rather than robot manipulation. To address this gap, we propose a framework to automatically design 3D printable adaptations that can be attached to hard-to-use objects, thus improving "robot ergonomics". Our learning-based framework formulates the adaptation design and control as a dual Markov decision process and is able to improve robot-object interactions for various robot end effectors and objects. We further validate our designs in the real world with a Franka Panda robot. Please see the supplementary video and https://object-adaptation.github.io for additional visualizations. Michelle Guo, Ziang Liu 0008, Stephen Tian, Zhaoming Xie, Jiajun Wu 0001, C. Karen Liu	ICRA	5
2024	Open X-Embodiment: Robotic Learning Datasets and RT-X Models : Open X-Embodiment Collaboration abstract Large, high-capacity models trained on diverse datasets have shown remarkable successes on efficiently tackling downstream applications. In domains from NLP to Computer Vision, this has led to a consolidation of pretrained models, with general pretrained backbones serving as a starting point for many applications. Can such a consolidation happen in robotics? Conventionally, robotic learning methods train a separate model for every application, every robot, and even every environment. Can we instead train "generalist" X-robot policy that can be adapted efficiently to new robots, tasks, and environments? In this paper, we provide datasets in standardized data formats and models to make it possible to explore this possibility in the context of robotic manipulation, alongside experimental results that provide an example of effective X-robot policies. We assemble a dataset from 22 different robots collected through a collaboration between 21 institutions, demonstrating 527 skills (160266 tasks). We show that a high-capacity model trained on this data, which we call RT-X, exhibits positive transfer and improves the capabilities of multiple robots by leveraging experience from other platforms. The project website is robotics-transformer-x.github.io. Abigail O'Neill, Abhiram Maddukuri, Abhishek Gupta 0004, Abhishek Padalkar, Abraham Lee, Acorn Pooley, Agrim Gupta, Ajay Mandlekar, Ajinkya Jain, Albert Tung, Alex Bewley, Alex Irpan, Alexander Khazatsky, Anant Rai, Anchit Gupta, Andrew E. Wang, Anikait Singh, Animesh Garg, Aniruddha Kembhavi, Annie Xie, Anthony Brohan, Antonin Raffin, Archit Sharma, Arefeh Yavary, Arhan Jain, Ashwin Balakrishna, Ayzaan Wahid, Ben Burgess-Limerick, Bernhard Schölkopf, Blake Wulfe, Brian Ichter, Cewu Lu, Charles Xu 0003, Charlotte Le, Chelsea Finn, Chen Wang 0053, Chenfeng Xu, Cheng Chi 0001, Chenguang Huang, Christine Chan, Christopher Agia, Chuer Pan, Chuyuan Fu, Coline Devin, Danfei Xu, Daniel Morton, Danny Drieß, Daphne Chen, Deepak Pathak, Dhruv Shah, Dieter Büchler, Dinesh Jayaraman, Dmitry Kalashnikov, Dorsa Sadigh, Edward Johns, Ethan Paul Foster, Fangchen Liu, Federico Ceola, Fei Xia 0002, Feiyu Zhao, Freek Stulp, Gaoyue Zhou, Gaurav S. Sukhatme, Gautam Salhotra, Gilbert Feng, Giulio Schiavi, Glen Berseth, Gregory Kahn, Guanzhi Wang, Hao Su 0001, Haoshu Fang, Henghui Bao, Heni Ben Amor, Henrik I. Christensen, Hiroki Furuta, Homer Walke, Hongjie Fang, Huy Ha, Igor Mordatch, Ilija Radosavovic, Isabel Leal, Jacky Liang, Jad Abou-Chakra, Jaehyung Kim 0001, Jaimyn Drake, Jan Peters 0001, Jan Schneider 0007, Jasmine Hsu, Jeannette Bohg, Jeffrey T. Bingham, Jensen Gao, Jiaheng Hu, Jiajun Wu 0001, Jiankai Sun, Jianlan Luo, Jiayuan Gu, Jie Tan 0001, Jihoon Oh, Jimmy Wu, Jingpei Lu, Jitendra Malik, João Silvério, Joey Hejna, Jonathan Booher, Jonathan Tompson, Jonathan Yang, Jordi Salvador, Joseph J. Lim, Junhyek Han, Kanishka Rao, Karl Pertsch, Karol Hausman, Keegan Go, Keerthana Gopalakrishnan, Kenneth Y. Goldberg, Kendra Byrne, Kenneth Oslund, Kento Kawaharazuka, Kevin Black, Kevin Zhang 0002, Kiana Ehsani, Kiran Lekkala, Kirsty Ellis, Krishan Rana, Krishnan Srinivasan, Kuan Fang, Kunal Pratap Singh, Kuo-Hao Zeng, Kyle Hatch, Kyle Hsu, Laurent Itti, Yunliang Chen 0001, Lerrel Pinto, Li Fei-Fei 0001, Liam Tan, Linxi Fan, Lionel Ott, Lisa Lee, Luca Weihs, Magnum Chen, Marion Lepert, Marius Memmel, Masayoshi Tomizuka, Masha Itkina, Mateo Guaman Castro, Max Spero, Maximilian Du, Michael Ahn, Michael C. Yip, Mingtong Zhang 0003, Mingyu Ding, Minho Heo, Mohan Kumar Srirama, Mohit Sharma 0001, Moo Jin Kim, Naoaki Kanazawa, Nicklas Hansen 0001, Nicolas Heess, Nikhil J. Joshi, Niko Sünderhauf, Norman Di Palo, Nur Muhammad Shafiullah, Oier Mees, Oliver Kroemer, Osbert Bastani, Pannag R. Sanketi, Patrick Tree Miller, Patrick Yin, Paul Wohlhart, Peng Xu 0010, Peter David Fagan, Peter Mitrano, Pierre Sermanet, Pieter Abbeel, Priya Sundaresan, Qiuyu Chen, Rafael Rafailov, Ria Doshi, Roberto Martin Martin, Rohan Baijal, Rosario Scalise, Rose Hendrix, Roy Lin, Runjia Qian, Russell Mendonca, Rutav Shah, Ryan Hoque, Ryan Julian, Samuel Bustamante-Gomez, Sean Kirmani, Sergey Levine, Sherry Moore, Shikhar Bahl, Shivin Dass, Shubham D. Sonawani, Shuran Song, Sichun Xu, Siddhant Haldar, Siddharth Karamcheti, Simeon Adebola, Simon Guist, Soroush Nasiriany, Stefan Schaal, Stefan Welker, Stephen Tian, Subramanian Ramamoorthy, Sudeep Dasari, Suneel Belkhale, Sungjae Park, Suraj Nair 0003, Suvir Mirchandani, Takayuki Osa, Tanmay Gupta, Tatsuya Harada, Tatsuya Matsushima, Ted Xiao, Thomas Kollar, Tianhe Yu, Tianli Ding, Todor Davchev, Tony Z. Zhao, Travis Armstrong, Trevor Darrell, Trinity Chung, Vidhi Jain, Vincent Vanhoucke, Wolfram Burgard, Xiaolong Wang 0004, Xinghao Zhu, Xinyang Geng, Liangwei Xu, Yecheng Jason Ma 0001, Yejin Kim 0003, Yevgen Chebotar, Yilin Wu 0003, Yonatan Bisk, Yoonyoung Cho, Youngwoon Lee, Yuchen Cui, Yueh-Hua Wu, Yujin Tang, Yuke Zhu, Yunchu Zhang, Yunfan Jiang 0001, Yunshuang Li, Yunzhu Li, Yusuke Iwasawa, Yutaka Matsuo, Zehan Ma, Zichen Jeff Cui, Zichen Zhang 0016, Zipeng Lin	ICRA	99
2024	HourVideo: 1-Hour Video-Language Understanding abstract We present HourVideo, a benchmark dataset for hour-long video-language understanding. Our dataset consists of a novel task suite comprising summarization, perception (recall, tracking), visual reasoning (spatial, temporal, predictive, causal, counterfactual), and navigation (room-to-room, object retrieval) tasks. HourVideo includes 500 manually curated egocentric videos from the Ego4D dataset, spanning durations of 20 to 120 minutes, and features 12,976 high-quality, five-way multiple-choice questions. Benchmarking results reveal that multimodal models, including GPT-4 and LLaVA-NeXT, achieve marginal improvements over random chance. In stark contrast, human experts significantly outperform the state-of-the-art long-context multimodal model, Gemini Pro 1.5 (85.0\% vs. 37.3\%), highlighting a substantial gap in multimodal capabilities. Our benchmark, evaluation toolkit, prompts, and documentation are available at https://hourvideo.stanford.edu. Keshigeyan Chandrasegaran, Agrim Gupta, Lea M. Hadzic, Taran Kota, Jimming He, Cristóbal Eyzaguirre, Zane Durante, Manling Li, Jiajun Wu 0001, Li Fei-Fei 0001	NeurIPS	9
2024	Streaming Detection of Queried Event Start abstract Robotics, autonomous driving, augmented reality, and many embodied computer vision applications must quickly react to user-defined events unfolding in real time. We address this setting by proposing a novel task for multimodal video understanding---Streaming Detection of Queried Event Start (SDQES).The goal of SDQES is to identify the beginning of a complex event as described by a natural language query, with high accuracy and low latency. We introduce a new benchmark based on the Ego4D dataset, as well as new task-specific metrics to study streaming multimodal detection of diverse events in an egocentric video setting.Inspired by parameter-efficient fine-tuning methods in NLP and for video tasks, we propose adapter-based baselines that enable image-to-video transfer learning, allowing for efficient online video modeling.We evaluate three vision-language backbones and three adapter architectures on both short-clip and untrimmed video settings. Cristóbal Eyzaguirre, Eric Tang, Shyamal Buch, Adrien Gaidon, Jiajun Wu 0001, Juan Carlos Niebles	NeurIPS	5
2024	MARPLE: A Benchmark for Long-Horizon Inference abstract Reconstructing past events requires reasoning across long time horizons. To figure out what happened, humans draw on prior knowledge about the world and human behavior and integrate insights from various sources of evidence including visual, language, and auditory cues. We introduce MARPLE, a benchmark for evaluating long-horizon inference capabilities using multi-modal evidence. Our benchmark features agents interacting with simulated households, supporting vision, language, and auditory stimuli, as well as procedurally generated environments and agent behaviors. Inspired by classic ``whodunit'' stories, we ask AI models and human participants to infer which agent caused a change in the environment based on a step-by-step replay of what actually happened. The goal is to correctly identify the culprit as early as possible. Our findings show that human participants outperform both traditional Monte Carlo simulation methods and an LLM baseline (GPT-4) on this task. Compared to humans, traditional inference models are less robust and performant, while GPT-4 has difficulty comprehending environmental changes. We analyze factors influencing inference performance and ablate different modes of evidence, finding that all modes are valuable for performance. Overall, our experiments demonstrate that the long-horizon, multimodal inference tasks in our benchmark present a challenge to current models. Project website: https://marple-benchmark.github.io/. Emily Jin, Zhuoyi Huang, Jan-Philipp Fränken, Hannah Cha, Erik Brockbank, Sarah A. Wu, Jiajun Wu 0001, Tobias Gerstenberg	NeurIPS	9
2024	Embodied Agent Interface: Benchmarking LLMs for Embodied Decision Making abstract We aim to evaluate Large Language Models (LLMs) for embodied decision making. While a significant body of work has been leveraging LLMs for decision making in embodied environments, we still lack a systematic understanding of their performance because they are usually applied in different domains, for different purposes, and built based on different inputs and outputs. Furthermore, existing evaluations tend to rely solely on a final success rate, making it difficult to pinpoint what ability is missing in LLMs and where the problem lies, which in turn blocks embodied agents from leveraging LLMs effectively and selectively. To address these limitations, we propose a generalized interface (Embodied Agent Interface) that supports the formalization of various types of tasks and input-output specifications of LLM-based modules. Specifically, it allows us to unify 1) a broad set of embodied decision-making tasks involving both state and temporally extended goals, 2) four commonly-used LLM-based modules for decision making: goal interpretation, subgoal decomposition, action sequencing, and transition modeling, and 3) a collection of fine-grained metrics that break down evaluation into error types, such as hallucination errors, affordance errors, and various types of planning errors. Overall, our benchmark offers a comprehensive assessment of LLMs’ performance for different subtasks, pinpointing the strengths and weaknesses in LLM-powered embodied AI systems and providing insights into the effective and selective use of LLMs in embodied decision making. Manling Li, Qineng Wang, Kangrui Wang, Sanjana Srivastava, Cem Gökmen, Li Erran Li, Percy Liang, Li Fei-Fei 0001, Jiayuan Mao, Jiajun Wu 0001	NeurIPS	15
2024	IKEA Manuals at Work: 4D Grounding of Assembly Instructions on Internet Videos abstract Shape assembly is a ubiquitous task in daily life, integral for constructing complex 3D structures like IKEA furniture. While significant progress has been made in developing autonomous agents for shape assembly, existing datasets have not yet tackled the 4D grounding of assembly instructions in videos, essential for a holistic understanding of assembly in 3D space over time. We introduce IKEA Video Manuals, a dataset that features 3D models of furniture parts, instructional manuals, assembly videos from the Internet, and most importantly, annotations of dense spatio-temporal alignments between these data modalities. To demonstrate the utility of IKEA Video Manuals, we present five applications essential for shape assembly: assembly plan generation, part-conditioned segmentation, part-conditioned pose estimation, video object segmentation, and furniture assembly based on instructional video manuals. For each application, we provide evaluation metrics and baseline methods. Through experiments on our annotated data, we highlight many challenges in grounding assembly instructions in videos to improve shape assembly, including handling occlusions, varying viewpoints, and extended assembly sequences. Yunong Liu, Cristóbal Eyzaguirre, Manling Li, Shubh Khanna, Juan Carlos Niebles, Vineeth Ravi, Saumitra Mishra, Jiajun Wu 0001	NeurIPS	9
2024	FactorSim: Generative Simulation via Factorized Representation abstract Generating simulations to train intelligent agents in game-playing and robotics from natural language input, user input, or task documentation remains an open-ended challenge. Existing approaches focus on parts of this challenge, such as generating reward functions or task hyperparameters. Unlike previous work, we introduce FACTORSIM that generates full simulations in code from language input that can be used to train agents. Exploiting the structural modularity specific to coded simulations, we propose to use a factored partially observable Markov decision process representation that allows us to reduce context dependence during each step of the generation. For evaluation, we introduce a generative simulation benchmark that assesses the generated simulation code’s accuracy and effectiveness in facilitating zero-shot transfers in reinforcement learning settings. We show that FACTORSIM outperforms existing methods in generating simulations regarding prompt alignment (i.e., accuracy), zero-shot transfer abilities, and human evaluation. We also demonstrate its effectiveness in generating robotic tasks. Fan-Yun Sun, S. I. Harini, Angela Yi, Alexander Zook, Jonathan Tremblay, Logan Matthew Cross, Jiajun Wu 0001, Nick Haber	NeurIPS	8
2024	An Eulerian Vortex Method on Flow Maps abstract We present an Eulerian vortex method based on the theory of flow maps to simulate the complex vortical motions of incompressible fluids. Central to our method is the novel incorporation of the flow-map transport equations for line elements , which, in combination with a bi-directional marching scheme for flow maps, enables the high-fidelity Eulerian advection of vorticity variables. The fundamental motivation is that, compared to impulse m , which has been recently bridged with flow maps to encouraging results, vorticity ω promises to be preferable for its numerical stability and physical interpretability. To realize the full potential of this novel formulation, we develop a new Poisson solving scheme for vorticity-to-velocity reconstruction that is both efficient and able to accurately handle the coupling near solid boundaries. We demonstrate the efficacy of our approach with a range of vortex simulation examples, including leapfrog vortices, vortex collisions, cavity flow, and the formation of complex vortical structures due to solid-fluid interactions. Yitong Deng, Molin Deng, Hong-Xing Yu, Junwei Zhou 0001, Duowen Chen 0003, Taku Komura, Jiajun Wu 0001, Bo Zhu 0002	ACM Trans. Graph.	8
2023	Learning Rational Subgoals from Demonstrations and Instructions abstract We present a framework for learning useful subgoals that support efficient long-term planning to achieve novel goals. At the core of our framework is a collection of rational subgoals (RSGs), which are essentially binary classifiers over the environmental states. RSGs can be learned from weakly-annotated data, in the form of unsegmented demonstration trajectories, paired with abstract task descriptions, which are composed of terms initially unknown to the agent (e.g., collect-wood then craft-boat then go-across-river). Our framework also discovers dependencies between RSGs, e.g., the task collect-wood is a helpful subgoal for the task craft-boat. Given a goal description, the learned subgoals and the derived dependencies facilitate off-the-shelf planning algorithms, such as A* and RRT, by setting helpful subgoals as waypoints to the planner, which significantly improves performance-time efficiency. Project page: https://rsg.csail.mit.edu Zhezheng Luo, Jiayuan Mao, Jiajun Wu 0001, Tomás Lozano-Pérez, Josh Tenenbaum, Leslie Pack Kaelbling	AAAI	3
2023	Learning to See the Physical World abstract This paper is part of the AAAI-23 New Faculty Highlights. In my presentation, I will introduce my research goal, which is to build machines that see, interact with, and reason about the physical world just like humans. This problem, which we call physical scene understanding, involves three key topics that bridge research in computer science, AI, robotics, cognitive science, and neuroscience: Perception, Physical Interaction, and Reasoning. Jiajun Wu 0001	AAAI	1
2023	Quantifying the Effect of Visual Impairments on Daily Activities in Virtual, Interactive Environments Wensi Ai, Sharon Lee, Li Fei-Fei 0001, Jiajun Wu 0001	CogSci	4
2023	Intuitions about physical scenes and objects in Virtual Reality (VR) Peter Zhu, Kanishk Gandhi, Yixing Wang, Steven Le, Jiajun Wu 0001, Hyowon Gweon	CogSci	5
2023	CIRCLE: Capture In Rich Contextual Environments abstract Synthesizing 3D human motion in a contextual, ecological environment is important for simulating realistic activities people perform in the real world. However, conventional optics-based motion capture systems are not suited for simultaneously capturing human movements and complex scenes. The lack of rich contextual 3D human motion datasets presents a roadblock to creating high-quality generative human motion models. We propose a novel motion acquisition system in which the actor perceives and operates in a highly contextual virtual world while being motion captured in the real world. Our system enables rapid collection of high-quality human motion in highly diverse scenes, without the concern of occlusion or the need for physical scene construction in the real world. We present CIRCLE, a dataset containing 10 hours of full-body reaching motion from 5 subjects across nine scenes, paired with ego-centric information of the environment represented in various forms, such as RGBD videos. We use this dataset to train a model that generates human motion conditioned on scene information. Leveraging our dataset, the model learns to use ego-centric scene information to achieve non-trivial reaching tasks in the context of complex 3D scenes. To download the data please visit our website. João Pedro Araújo 0001, Jiaman Li, Karthik Vetrivel, Rishi Agarwal, Jiajun Wu 0001, Deepak Gopinath, Alexander Clegg, C. Karen Liu	CVPR	5
2023	REALIMPACT: A Dataset of Impact Sound Fields for Real Objects abstract Objects make unique sounds under different perturbations, environment conditions, and poses relative to the listener. While prior works have modeled impact sounds and sound propagation in simulation, we lack a standard dataset of impact sound fields of real objects for audio-visual learning and calibration of the sim-to-real gap. We present Realimpact,a large-scale dataset of real object impact sounds recorded under controlled conditions. Re-alimpactcontains 150,000 recordings of impact sounds of 50 everyday objects with detailed annotations, including their impact locations, microphone locations, contact force profiles, material labels, and RGBD images.**The project page and dataset are available at https://samuelpclarke.com/realimpact/ We make preliminary attempts to use our dataset as a reference to current simulation methods for estimating object impact sounds that match the real world. Moreover, we demon-strate the usefulness of our dataset as a testbed for acoustic and audio-visual learning via the evaluation of two bench-mark tasks, including listener location classification and vi-sual acoustic matching. Samuel Clarke, Ruohan Gao, Mason L. Wang, Mark Rau, Julia Xu, Jui-Hsien Wang, Doug L. James, Jiajun Wu 0001	CVPR	8
2023	The Object Folder Benchmark : Multisensory Learning with Neural and Real Objects abstract We introduce the ObjectFolder Benchmark, a benchmark suite of 10 tasks for multisensory object-centric learning, centered around object recognition, reconstruction, and manipulation with sight, sound, and touch. We also introduce the Objectfolder Real dataset, including the multisensory measurements for 100 real-world household objects, building upon a newly designed pipeline for collecting the 3D meshes, videos, impact sounds, and tactile readings of real-world objects. We conduct systematic benchmarking on both the 1,000 multisensory neural objects from Objectfolder, and the real multisensory data from Objectfolder Real. Our results demonstrate the importance of multisensory perception and reveal the respective roles of vision, audio, and touch for different object-centric learning tasks. By publicly releasing our dataset and benchmark suite, we hope to catalyze and enable new research in multisensory object-centric learning in computer vision, robotics, and beyond. Project page: https://objectfolder.stanford.edu Ruohan Gao, Yiming Dou, Hao Li 0076, Tanmay Agarwal, Jeannette Bohg, Yunzhu Li, Li Fei-Fei 0001, Jiajun Wu 0001	CVPR	8
2023	NS3D: Neuro-Symbolic Grounding of 3D Objects and Relations abstract Grounding object properties and relations in 3D scenes is a prerequisite for a wide range of artificial intelligence tasks, such as visually grounded dialogues and embodied manipulation. However, the variability of the 3D domain induces two fundamental challenges: 1) the expense of labeling and 2) the complexity of 3D grounded language. Hence, essential desiderata for models are to be data-efficient, generalize to different data distributions and tasks with unseen semantic forms, as well as ground complex language semantics (e.g., view-point anchoring and multi-object reference). To address these challenges, we propose NS3D, a neuro-symbolic framework for 3D grounding. NS3D translates language into programs with hierarchical structures by leveraging large language-to-code models. Different functional modules in the programs are implemented as neural networks. Notably, NS3D extends prior neuro-symbolic visual reasoning methods by introducing functional modules that effectively reason about high-arity relations (i.e., relations among more than two objects), key in disambiguating objects in complex 3D scenes. Modular and compositional architecture enables NS3D to achieve state-of-the-art results on the ReferIt3D view-dependence task, a 3D referring expression compre-hension benchmark. Importantly, NS3D shows significantly improved performance on settings of data-efficiency and generalization, and demonstrate zero-shot transfer to an unseen 3D question-answering task. Joy Hsu, Jiayuan Mao, Jiajun Wu 0001	CVPR	3
2023	Putting People in Their Place: Affordance-Aware Human Insertion into Scenes abstract We study the problem of inferring scene affordances by presenting a method for realistically inserting people into scenes. Given a scene image with a marked region and an image of a person, we insert the person into the scene while respecting the scene affordances. Our model can infer the set of realistic poses given the scene context, re-pose the reference person, and harmonize the composition. We set up the task in a self-supervised fashion by learning to repose humans in video clips. We train a large-scale diffusion model on a dataset of 2.4M video clips that produces diverse plausible poses while respecting the scene context. Given the learned human-scene composition, our model can also hallucinate realistic people and scenes when prompted without conditioning and also enables interactive editing. A quantitative evaluation shows that our method synthesizes more realistic human appearance and more natural human-scene interactions than prior work. Sumith Kulal, Tim Brooks, Alex Aiken, Jiajun Wu 0001, Jimei Yang, Jingwan Lu, Alexei A. Efros, Krishna Kumar Singh	CVPR	4
2023	Ego-Body Pose Estimation via Ego-Head Pose Estimation abstract Estimating 3D human motion from an egocentric video sequence plays a critical role in human behavior understanding and has various applications in VR/AR. However, naively learning a mapping between egocentric videos and human motions is challenging, because the user's body is often unobserved by the front-facing camera placed on the head of the user. In addition, collecting large-scale, high-quality datasets with paired egocentric videos and 3D human motions requires accurate motion capture devices, which often limit the variety of scenes in the videos to lab-like environments. To eliminate the need for paired egocentric video and human motions, we propose a new method, Ego-Body Pose Estimation via Ego-Head Pose Estimation (EgoEgo), which decomposes the problem into two stages, connected by the head motion as an intermediate representation. EgoEgo first integrates SLAM and a learning approach to estimate accurate head motion. Subsequently, leveraging the estimated head pose as input, EgoEgo utilizes conditional diffusion to generate multiple plausible full-body motions. This disentanglement of head and body pose eliminates the need for training datasets with paired egocentric videos and 3D human motion, enabling us to leverage large-scale egocentric video datasets and motion capture datasets separately. Moreover, for systematic benchmarking, we develop a synthetic dataset, AMASS-Replica-Ego-Syn (ARES), with paired egocentric videos and human motion. On both ARES and real data, our EgoEgo model performs significantly better than the current state-of-the-art methods. Jiaman Li, C. Karen Liu, Jiajun Wu 0001	CVPR	3
2023	3D Neural Field Generation Using Triplane Diffusion abstract Diffusion models have emerged as the state-of-the-art for image generation, among other tasks. Here, we present an efficient diffusion-based model for 3D-aware generation of neural fields. Our approach pre-processes training data, such as ShapeNet meshes, by converting them to continuous occupancy fields and factoring them into a set of axis-aligned triplane feature representations. Thus, our 3D training scenes are all represented by 2D feature planes, and we can directly train existing 2D diffusion models on these representations to generate 3D neural fields with high quality and diversity, outperforming alternative approaches to 3D-aware generation. Our approach requires essential modifications to existing triplane factorization pipelines to make the resulting features easy to learn for the diffusion model. We demonstrate state-of-the-art results on 3D generation on several object classes from ShapeNet. J. Ryan Shue, Eric R. Chan, Ryan Po, Zachary Ankner, Jiajun Wu 0001, Gordon Wetzstein	CVPR	5
2023	Multi-Object Manipulation via Object-Centric Neural Scattering Functions abstract Learned visual dynamics models have proven effective for robotic manipulation tasks. Yet, it remains unclear how best to represent scenes involving multi-object interactions. Current methods decompose a scene into discrete objects, but they struggle with precise modeling and manipulation amid challenging lighting conditions as they only encode appearance tied with specific illuminations. In this work, we propose using object-centric neural scattering functions (OSFs) as object representations in a model-predictive control framework. OSFs model per-object light transport, enabling compositional scene re-rendering under object rearrangement and varying lighting conditions. By combining this approach with inverse parameter estimation and graph-based neural dynamics models, we demonstrate improved model-predictive control performance and generalization in compositional multi-object environments, even in previously unseen scenarios and harsh lighting conditions. Stephen Tian, Yancheng Cai, Hong-Xing Yu, Sergey Zakharov, Katherine Liu, Adrien Gaidon, Yunzhu Li, Jiajun Wu 0001	CVPR	8
2023	PyPose: A Library for Robot Learning with Physics-based Optimization abstract Deep learning has had remarkable success in robotic perception, but its data-centric nature suffers when it comes to generalizing to ever-changing environments. By contrast, physics-based optimization generalizes better, but it does not perform as well in complicated tasks due to the lack of high-level semantic information and reliance on manual parametric tuning. To take advantage of these two complementary worlds, we present PyPose: a robotics-oriented, PyTorch-based library that combines deep perceptual models with physics-based optimization. PyPose's architecture is tidy and well-organized, it has an imperative style interface and is efficient and user-friendly, making it easy to integrate into real-world robotic applications. Besides, it supports parallel computing of any order gradients of Lie groups and Lie algebras and 2nd-order optimizers, such as trust region methods. Experiments show that PyPose achieves more than 10× speedup in computation compared to the state-of-the-art libraries. To boost future research, we provide concrete examples for several fields of robot learning, including SLAM, planning, control, and inertial navigation. Chen Wang 0033, Dasong Gao, Junyi Geng, Yaoyu Hu, Yuheng Qiu, Bowen Li 0007, Fan Yang 0092, Brady G. Moon, Abhinav Pandey, Aryan, Jiahe Xu 0002, Daning Huang, Zhongqiang Ren, Shibo Zhao, Taimeng Fu, Pranay Reddy, Jingnan Shi, Rajat Talak, Kun Cao 0002, Yi Du 0001, Huai Yu, Shanzhao Wang, Siyu Chen 0036, Ananth Kashyap, Rohan Bandaru, Karthik Dantu, Jiajun Wu 0001, Lihua Xie 0001, Luca Carlone, Marco Hutter 0001, Sebastian A. Scherer	CVPR	33
2023	ULIP: Learning a Unified Representation of Language, Images, and Point Clouds for 3D Understanding abstract The recognition capabilities of current state-of-the-art 3D models are limited by datasets with a small number of annotated data and a pre-defined set of categories. In its 2D counterpart, recent advances have shown that similar problems can be significantly alleviated by employing knowledge from other modalities, such as language. Inspired by this, leveraging multimodal information for 3D modality could be promising to improve 3D understanding under the restricted data regime, but this line of research is not well studied. Therefore, we introduce ULIP to learn a unified representation of image, text, and 3D point cloud by pre-training with object triplets from the three modalities. To overcome the shortage of training triplets, ULIP leverages a pre-trained vision-language model that has already learned a common visual and textual space by training with massive image-text pairs. Then, ULIP learns a 3D representation space aligned with the common image-text space, using a small number of automatically synthesized triplets. ULIP is agnostic to 3D backbone networks and can easily be integrated into any 3D architecture. Experiments show that ULIP effectively improves the performance of multiple recent 3D backbones by simply pre-training them on ShapeNet55 using our framework, achieving state-of-the-art performance in both standard 3D classification and zero-shot 3D classification on ModelNet40 and ScanObjectNN. ULIP also improves the performance of PointMLP by around 3% in 3D classification on ScanObjectNN, and outperforms PointCLIP by 28.8% on top-1 accuracy for zero-shot 3D classification on ModelNet40. Our code and pre-trained models will be released. Le Xue, Mingfei Gao, Chen Xing, Roberto Martin Martin, Jiajun Wu 0001, Caiming Xiong, Ran Xu 0001, Juan Carlos Niebles, Silvio Savarese	CVPR	5
2023	Accidental Light Probes abstract Recovering lighting in a scene from a single image is a fundamental problem in computer vision. While a mirror ball light probe can capture omnidirectional lighting, light probes are generally unavailable in everyday images. In this work, we study recovering lighting from accidental light probes (ALPs)-common, shiny objects like Coke cans, which often accidentally appear in daily scenes. We propose a physically-based approach to model ALPs and estimate lighting from their appearances in single images. The main idea is to model the appearance of ALPs by photogram-metrically principled shading and to invert this process via differentiable rendering to recover incidental illumination. We demonstrate that we can put an ALP into a scene to allow high-fidelity lighting estimation. Our model can also recover lighting for existing images that happen to contain an ALP**Project website: https://kovenyu.com/ALP. I'd rather be Shiny. - Tamatoa from Moana, 2016 Hong-Xing Yu, Samir Agarwala, Charles Herrmann, Richard Szeliski, Noah Snavely, Jiajun Wu 0001, Deqing Sun	CVPR	6
2023	Seeing a Rose in Five Thousand Ways abstract What is a rose, visually? A rose comprises its intrinsics, including the distribution of geometry, texture, and material specific to its object category. With knowledge of these intrinsic properties, we may render roses of different sizes and shapes, in different poses, and under different lighting conditions. In this work, we build a generative model that learns to capture such object intrinsics from a single image, such as a photo of a bouquet. Such an image includes multiple instances of an object type. These instances all share the same intrinsics, but appear different due to a combination of variance within these intrinsics and differences in extrinsic factors, such as pose and illumination. Experiments show that our model successfully learns object intrinsics (distribution of geometry, texture, and material) for a wide range of objects, each from a single Internet image. Our method achieves superior results on multiple downstream tasks, including intrinsic image decomposition, shape and image generation, view synthesis, and relighting. “Nature! … Each of her works has an essence of its own; each of her phenomena a special characterisation; and yet their diversity is in unity.” - Georg Christoph Tobler Shangzhe Wu, Noah Snavely, Jiajun Wu 0001	CVPR	4
2023	Tree-Structured Shading Decomposition abstract We study inferring a tree-structured representation from a single image for object shading. Prior work typically uses the parametric or measured representation to model shading, which is neither interpretable nor easily editable. We propose using the shade tree representation, which combines basic shading nodes and compositing methods to factorize object surface shading. The shade tree representation enables novice users who are unfamiliar with the physical shading process to edit object shading in an efficient and intuitive manner. A main challenge in inferring the shade tree is that the inference problem involves both the discrete tree structure and the continuous parameters of the tree nodes. We propose a hybrid approach to address this issue. We introduce an auto-regressive inference model to generate a rough estimation of the tree structure and node parameters, and then we fine-tune the inferred shade tree through an optimization algorithm. We show experiments on synthetic images, captured reflectance, real images, and non-realistic vector drawings, allowing downstream applications such as material editing, vectorized shading, and relighting. Project website: https://chen-geng.com/inv-shade-trees. Chen Geng 0001, Hong-Xing Yu, Sharon Zhang, Maneesh Agrawala, Jiajun Wu 0001	ICCV	5
2023	VQ3D: Learning a 3D-Aware Generative Model on ImageNet abstract Recent work has shown the possibility of training generative models of 3D content from 2D image collections on small datasets corresponding to a single object class, such as human faces, animal faces, or cars. However, these models struggle on larger, more complex datasets. To model diverse and unconstrained image collections such as ImageNet, we present VQ3D, which introduces a NeRF-based decoder into a two-stage vector-quantized autoencoder. Our Stage 1 allows for the reconstruction of an input image and the ability to change the camera position around the image, and our Stage 2 allows for the generation of new 3D scenes. VQ3D is capable of generating and reconstructing 3D-aware images from the 1000-class ImageNet dataset of 1.2 million training images, and achieves a competitive ImageNet generation FID score of 16.8. Our project webpage is at this url. Kyle Sargent, Jing Yu Koh, Han Zhang 0010, Huiwen Chang, Charles Herrmann, Pratul P. Srinivasan, Jiajun Wu 0001, Deqing Sun	ICCV	7
2023	Rendering Humans from Object-Occluded Monocular Videos abstract 3D understanding and rendering of moving humans from monocular videos is a challenging task. Despite recent progress, the task remains difficult in real-world scenarios, where obstacles may block the camera view and cause partial occlusions in the captured videos. Existing methods cannot handle such defects due to two reasons. First, the standard rendering strategy relies on point-point mapping, which could lead to dramatic disparities between the visible and occluded areas of the body. Second, the naive direct regression approach does not consider any feasibility criteria (i.e., prior information) for rendering under occlusions. To tackle the above drawbacks, we present OccNeRF, a neural rendering method that achieves better rendering of humans in severely occluded scenes. As direct solutions to the two drawbacks, we propose surface-based rendering by integrating geometry and visibility priors. We validate our method on both simulated and real-world occlusions and demonstrate our method’s superiority. Project page: https://cs.stanford.edu/~xtiange/projects/occnerf/ Tiange Xiang, Adam Sun, Jiajun Wu 0001, Ehsan Adeli-Mosabbeb, Li Fei-Fei 0001	ICCV	3
2023	Learning Vortex Dynamics for Fluid Inference and Prediction Yitong Deng, Hong-Xing Yu, Jiajun Wu 0001, Bo Zhu 0002	ICLR	3
2023	MaskViT: Masked Visual Pre-Training for Video Prediction Agrim Gupta, Stephen Tian, Jiajun Wu 0001, Roberto Martin Martin, Li Fei-Fei 0001	ICLR	4
2023	An Extensible Multi-modal Multi-task Object Dataset with Materials Trevor Scott Standley, Ruohan Gao, Dawn Chen, Jiajun Wu 0001, Silvio Savarese	ICLR	4
2023	A Control-Centric Benchmark for Video Prediction Stephen Tian, Chelsea Finn, Jiajun Wu 0001	ICLR	3
2023	Programmatically Grounded, Compositionally Generalizable Robotic Manipulation Renhao Wang, Jiayuan Mao, Joy Hsu, Hang Zhao 0021, Jiajun Wu 0001, Yang Gao 0029	ICLR	5
2023	Motion Question Answering via Modular Motion Programs abstract In order to build artificial intelligence systems that can perceive and reason with human behavior in the real world, we must first design models that conduct complex spatio-temporal reasoning over motion sequences. Moving towards this goal, we propose the HumanMotionQA task to evaluate complex, multi-step reasoning abilities of models on long-form human motion sequences. We generate a dataset of question-answer pairs that require detecting motor cues in small portions of motion sequences, reasoning temporally about when events occur, and querying specific motion attributes. In addition, we propose NSPose, a neuro-symbolic method for this task that uses symbolic reasoning and a modular design to ground motion through learning motion concepts, attribute neural operators, and temporal relations. We demonstrate the suitability of NSPose for the HumanMotionQA task, outperforming all baseline methods. Mark Endo, Joy Hsu, Jiaman Li, Jiajun Wu 0001	ICML	4
2023	Modeling Dynamic Environments with Scene Graph Memory abstract Embodied AI agents that search for objects in large environments such as households often need to make efficient decisions by predicting object locations based on partial information. We pose this as a new type of link prediction problem: link prediction on partially observable dynamic graphs Our graph is a representation of a scene in which rooms and objects are nodes, and their relationships are encoded in the edges; only parts of the changing graph are known to the agent at each timestep. This partial observability poses a challenge to existing link prediction approaches, which we address. We propose a novel state representation – Scene Graph Memory (SGM) – with captures the agent’s accumulated set of observations, as well as a neural net architecture called a Node Edge Predictor (NEP) that extracts information from the SGM to search efficiently. We evaluate our method in the Dynamic House Simulator, a new benchmark that creates diverse dynamic graphs following the semantic patterns typically seen at homes, and show that NEP can be trained to predict the locations of objects in a variety of environments with diverse object movement dynamics, outperforming baselines both in terms of new scene adaptability and overall accuracy. The codebase and more can be found www.scenegraphmemory.com. Andrey Kurenkov, Michael Lingelbach, Tanmay Agarwal, Emily Jin, Chengshu Li 0002, Li Fei-Fei 0001, Jiajun Wu 0001, Silvio Savarese, Roberto Martin Martin	ICML	8
2023	STAP: Sequencing Task-Agnostic Policies abstract Advances in robotic skill acquisition have made it possible to build general-purpose libraries of learned skills for downstream manipulation tasks. However, naively executing these skills one after the other is unlikely to succeed without accounting for dependencies between actions prevalent in longhorizon plans. We present Sequencing Task-Agnostic Policies (STAP), a scalable framework for training manipulation skills and coordinating their geometric dependencies at planning time to solve long-horizon tasks never seen by any skill during training. Given that Q-functions encode a measure of skill feasibility, we formulate an optimization problem to maximize the joint success of all skills sequenced in a plan, which we estimate by the product of their Q-values. Our experiments indicate that this objective function approximates ground truth plan feasibility and, when used as a planning objective, reduces myopic behavior and thereby promotes long-horizon task success. We further demonstrate how STAP can be used for task and motion planning by estimating the geometric feasibility of skill sequences provided by a task planner. We evaluate our approach in simulation and on a real robot. Qualitative results and code are made available at sites.google.com/stanford.edu/stap. Christopher Agia, Toki Migimatsu, Jiajun Wu 0001, Jeannette Bohg	ICRA	3
2023	Sonicverse: A Multisensory Simulation Platform for Embodied Household Agents that See and Hear abstract Developing embodied agents in simulation has been a key research topic in recent years. Exciting new tasks, algorithms, and benchmarks have been developed in various simulators. However, most of them assume deaf agents in silent environments, while we humans perceive the world with multiple senses. We introduce Sonicverse, a multisensory simulation platform with integrated audio-visual simulation for training household agents that can both see and hear. Sonicverse models realistic continuous audio rendering in 3D environments in real-time. Together with a new audio-visual VR interface that allows humans to interact with agents with audio, Sonicverse enables a series of embodied AI tasks that need audio-visual perception. For semantic audio-visual navigation in particular, we also propose a new multi-task learning model that achieves state-of-the-art performance. In addition, we demonstrate Sonicverse's realism via sim-to-real transfer, which has not been achieved by other simulators: an agent trained in Sonicverse can successfully perform audio-visual navigation in real-world environments. Sonicverse is available at: https://github.com/StanfordVL/Sonicverse. Ruohan Gao, Hao Li 0076, Gokul Dharan, Chengshu Li 0002, Fei Xia 0002, Silvio Savarese, Li Fei-Fei 0001, Jiajun Wu 0001	ICRA	9
2023	Task-Driven Graph Attention for Hierarchical Relational Object Navigation abstract Embodied AI agents in large scenes often need to navigate to find objects. In this work, we study a naturally emerging variant of the object navigation task, hierarchical relational object navigation (HRON), where the goal is to find objects specified by logical predicates organized in a hierarchical structure-objects related to furniture and then to rooms-such as finding an apple on top of a table in the kitchen. Solving such a task requires an efficient representation to reason about object relations and correlate the relations in the environment and in the task goal. HRON in large scenes (e.g. homes) is particularly challenging due to its partial observability and long horizon, which invites solutions that can compactly store the past information while effectively exploring the scene. We demonstrate experimentally that scene graphs are the best-suited representation compared to conventional representations such as images or 2D maps. We propose a solution that uses scene graphs as part of its input and integrates graph neural networks as its backbone, with an integrated task-driven attention mechanism, and demonstrate its better scalability and learning efficiency than state-of-the-art baselines. Michael Lingelbach, Chengshu Li 0002, Minjune Hwang, Andrey Kurenkov, Alan Lou, Roberto Martin Martin, Li Fei-Fei 0001, Jiajun Wu 0001	ICRA	9
2023	Primitive Skill-Based Robot Learning from Human Evaluative Feedback abstract Reinforcement learning (RL) algorithms face significant challenges when dealing with long-horizon robot manipulation tasks in real-world environments due to sample inefficiency and safety issues. To overcome these challenges, we propose a novel framework, SEED, which leverages two approaches: reinforcement learning from human feedback (RLHF) and primitive skill-based reinforcement learning. Both approaches are particularly effective in addressing sparse reward issues and the complexities involved in long-horizon tasks. By combining them, SEED reduces the human effort required in RLHF and increases safety in training robot manipulation with RL in real-world settings. Additionally, parameterized skills provide a clear view of the agent's high-level intentions, allowing humans to evaluate skill choices before they are executed. This feature makes the training process even safer and more efficient. To evaluate the performance of SEED, we conducted extensive experiments on five manipulation tasks with varying levels of complexity. Our results show that SEED significantly outperforms state-of-the-art RL algorithms in sample efficiency and safety. In addition, SEED also exhibits a substantial reduction of human effort compared to other RLHF methods. Further details and video results can be found at https://seediros23.github.io/. Ayano Hiranaka, Minjune Hwang, Sharon Lee, Chen Wang 0053, Li Fei-Fei 0001, Jiajun Wu 0001	IROS	6
2023	Model-Based Control with Sparse Neural Dynamics abstract Learning predictive models from observations using deep neural networks (DNNs) is a promising new approach to many real-world planning and control problems. However, common DNNs are too unstructured for effective planning, and current control methods typically rely on extensive sampling or local gradient descent. In this paper, we propose a new framework for integrated model learning and predictive control that is amenable to efficient optimization algorithms. Specifically, we start with a ReLU neural model of the system dynamics and, with minimal losses in prediction accuracy, we gradually sparsify it by removing redundant neurons. This discrete sparsification process is approximated as a continuous problem, enabling an end-to-end optimization of both the model architecture and the weight parameters. The sparsified model is subsequently used by a mixed-integer predictive controller, which represents the neuron activations as binary variables and employs efficient branch-and-bound algorithms. Our framework is applicable to a wide variety of DNNs, from simple multilayer perceptrons to complex graph neural dynamics. It can efficiently handle tasks involving complicated contact dynamics, such as object pushing, compositional object sorting, and manipulation of deformable objects. Numerical and hardware experiments show that, despite the aggressive sparsification, our framework can deliver better closed-loop performance than existing state-of-the-art methods. Ziang Liu 0008, Genggeng Zhou, Jeff He, Tobia Marcucci, Li Fei-Fei 0001, Jiajun Wu 0001, Yunzhu Li	NeurIPS	6
2023	3D Copy-Paste: Physically Plausible Object Insertion for Monocular 3D Detection abstract A major challenge in monocular 3D object detection is the limited diversity and quantity of objects in real datasets. While augmenting real scenes with virtual objects holds promise to improve both the diversity and quantity of the objects, it remains elusive due to the lack of an effective 3D object insertion method in complex real captured scenes. In this work, we study augmenting complex real indoor scenes with virtual objects for monocular 3D object detection. The main challenge is to automatically identify plausible physical properties for virtual assets (e.g., locations, appearances, sizes, etc.) in cluttered real scenes. To address this challenge, we propose a physically plausible indoor 3D object insertion approach to automatically copy virtual objects and paste them into real scenes. The resulting objects in scenes have 3D bounding boxes with plausible physical locations and appearances. In particular, our method first identifies physically feasible locations and poses for the inserted objects to prevent collisions with the existing room layout. Subsequently, it estimates spatially-varying illumination for the insertion location, enabling the immersive blending of the virtual objects into the original scene with plausible appearances and cast shadows. We show that our augmentation method significantly improves existing monocular 3D object models and achieves state-of-the-art performance. For the first time, we demonstrate that a physically plausible 3D object insertion, serving as a generative data augmentation technique, can lead to significant improvements for discriminative downstream tasks such as monocular 3D object detection. Project website: https://gyhandy.github.io/3D-Copy-Paste/. Yunhao Ge, Hong-Xing Yu, Cheng Zhao 0002, Yuliang Guo, Xinyu Huang 0001, Liu Ren 0001, Laurent Itti, Jiajun Wu 0001	NeurIPS	8
2023	Siamese Masked Autoencoders abstract Establishing correspondence between images or scenes is a significant challenge in computer vision, especially given occlusions, viewpoint changes, and varying object appearances. In this paper, we present Siamese Masked Autoencoders (SiamMAE), a simple extension of Masked Autoencoders (MAE) for learning visual correspondence from videos. SiamMAE operates on pairs of randomly sampled video frames and asymmetrically masks them. These frames are processed independently by an encoder network, and a decoder composed of a sequence of cross-attention layers is tasked with predicting the missing patches in the future frame. By masking a large fraction (95%) of patches in the future frame while leaving the past frame unchanged, SiamMAE encourages the network to focus on object motion and learn object-centric representations. Despite its conceptual simplicity, features learned via SiamMAE outperform state-of-the-art self-supervised methods on video object segmentation, pose keypoint propagation, and semantic part propagation tasks. SiamMAE achieves competitive results without relying on data augmentation, handcrafted tracking-based pretext tasks, or other techniques to prevent representational collapse. Agrim Gupta, Jiajun Wu 0001, Jia Deng 0001, Li Fei-Fei 0001	NeurIPS	2
2023	Disentanglement via Latent Quantization abstract In disentangled representation learning, a model is asked to tease apart a dataset's underlying sources of variation and represent them independently of one another. Since the model is provided with no ground truth information about these sources, inductive biases take a paramount role in enabling disentanglement. In this work, we construct an inductive bias towards encoding to and decoding from an organized latent space. Concretely, we do this by (i) quantizing the latent space into discrete code vectors with a separate learnable scalar codebook per dimension and (ii) applying strong model regularization via an unusually high weight decay. Intuitively, the latent space design forces the encoder to combinatorially construct codes from a small number of distinct scalar values, which in turn enables the decoder to assign a consistent meaning to each value. Regularization then serves to drive the model towards this parsimonious strategy. We demonstrate the broad applicability of this approach by adding it to both basic data-reconstructing (vanilla autoencoder) and latent-reconstructing (InfoGAN) generative models. For reliable evaluation, we also propose InfoMEC, a new set of metrics for disentanglement that is cohesively grounded in information theory and fixes well-established shortcomings in previous metrics. Together with regularization, latent quantization dramatically improves the modularity and explicitness of learned representations on a representative suite of benchmark datasets. In particular, our quantized-latent autoencoder (QLAE) consistently outperforms strong methods from prior work in these key disentanglement properties without compromising data reconstruction. Kyle Hsu, William Dorrell, James C. R. Whittington, Jiajun Wu 0001, Chelsea Finn	NeurIPS	4
2023	What's Left? Concept Grounding with Logic-Enhanced Foundation Models abstract Recent works such as VisProg and ViperGPT have smartly composed foundation models for visual reasoning—using large language models (LLMs) to produce programs that can be executed by pre-trained vision-language models. However, they operate in limited domains, such as 2D images, not fully exploiting the generalization of language: abstract concepts like “left” can also be grounded in 3D, temporal, and action data, as in moving to your left. This limited generalization stems from these inference-only methods’ inability to learn or adapt pre-trained models to a new domain. We propose the Logic-Enhanced FoundaTion Model (LEFT), a unified framework that learns to ground and reason with concepts across domains with a differentiable, domain-independent, first-order logic-based program executor. LEFT has an LLM interpreter that outputs a program represented in a general, logic-based reasoning language, which is shared across all domains and tasks. LEFT’s executor then executes the program with trainable domain-specific grounding modules. We show that LEFT flexibly learns concepts in four domains: 2D images, 3D scenes, human motions, and robotic manipulation. It exhibits strong reasoning ability in a wide variety of tasks, including those that are complex and not seen during training, and can be easily applied to new domains. Joy Hsu, Jiayuan Mao, Josh Tenenbaum, Jiajun Wu 0001	NeurIPS	4
2023	Are These the Same Apple? Comparing Images Based on Object Intrinsics abstract The human visual system can effortlessly recognize an object under different extrinsic factors such as lighting, object poses, and background, yet current computer vision systems often struggle with these variations. An important step to understanding and improving artificial vision systems is to measure image similarity purely based on intrinsic object properties that define object identity. This problem has been studied in the computer vision literature as re-identification, though mostly restricted to specific object categories such as people and cars. We propose to extend it to general object categories, exploring an image similarity metric based on object intrinsics. To benchmark such measurements, we collect the Common paired objects Under differenT Extrinsics (CUTE) dataset of 18, 000 images of 180 objects under different extrinsic factors such as lighting, poses, and imaging conditions. While existing methods such as LPIPS and CLIP scores do not measure object intrinsics well, we find that combining deep features learned from contrastive self-supervised learning with foreground filtering is a simple yet effective approach to approximating the similarity. We conduct an extensive survey of pre-trained features and foreground extraction methods to arrive at a strong baseline that best measures intrinsic object-centric image similarity among current methods. Finally, we demonstrate that our approach can aid in downstream applications such as acting as an analog for human subjects and improving generalizable re-identification. Please see our project website at https://s-tian.github.io/projects/cute/ for visualizations of the data and demos of our metric. Klemen Kotar, Stephen Tian, Hong-Xing Yu, Dan Yamins, Jiajun Wu 0001	NeurIPS	5
2023	Stanford-ORB: A Real-World 3D Object Inverse Rendering Benchmark abstract We introduce Stanford-ORB, a new real-world 3D Object inverse Rendering Benchmark. Recent advances in inverse rendering have enabled a wide range of real-world applications in 3D content generation, moving rapidly from research and commercial use cases to consumer devices. While the results continue to improve, there is no real-world benchmark that can quantitatively assess and compare the performance of various inverse rendering methods. Existing real-world datasets typically only consist of the shape and multi-view images of objects, which are not sufficient for evaluating the quality of material recovery and object relighting. Methods capable of recovering material and lighting often resort to synthetic data for quantitative evaluation, which on the other hand does not guarantee generalization to complex real-world environments. We introduce a new dataset of real-world objects captured under a variety of natural scenes with ground-truth 3D scans, multi-view images, and environment lighting. Using this dataset, we establish the first comprehensive real-world evaluation benchmark for object inverse rendering tasks from in-the-wild scenes, and compare the performance of various existing methods. All data, code, and models can be accessed at https://stanfordorb.github.io/ Zhengfei Kuang, Hong-Xing Yu, Samir Agarwala, Shangzhe Wu, Jiajun Wu 0001	NeurIPS	6
2023	Holistic Evaluation of Text-to-Image Models abstract The stunning qualitative improvement of text-to-image models has led to their widespread attention and adoption. However, we lack a comprehensive quantitative understanding of their capabilities and risks. To fill this gap, we introduce a new benchmark, Holistic Evaluation of Text-to-Image Models (HEIM). Whereas previous evaluations focus mostly on image-text alignment and image quality, we identify 12 aspects, including text-image alignment, image quality, aesthetics, originality, reasoning, knowledge, bias, toxicity, fairness, robustness, multilinguality, and efficiency. We curate 62 scenarios encompassing these aspects and evaluate 26 state-of-the-art text-to-image models on this benchmark. Our results reveal that no single model excels in all aspects, with different models demonstrating different strengths. We release the generated images and human evaluation results for full transparency at https://crfm.stanford.edu/heim/latest and the code at https://github.com/stanford-crfm/helm, which is integrated with the HELM codebase Michihiro Yasunaga, Chenlin Meng, Yifan Mai 0001, Joon Sung Park 0001, Agrim Gupta, Deepak Narayanan, Hannah Teufel, Marco Bellagente, Minguk Kang, Taesung Park, Jure Leskovec, Jun-Yan Zhu, Li Fei-Fei 0001, Jiajun Wu 0001, Stefano Ermon, Percy Liang	NeurIPS	16
2023	SoundCam: A Dataset for Finding Humans Using Room Acoustics abstract A room’s acoustic properties are a product of the room’s geometry, the objects within the room, and their specific positions. A room’s acoustic properties can be characterized by its impulse response (RIR) between a source and listener location, or roughly inferred from recordings of natural signals present in the room. Variations in the positions of objects in a room can effect measurable changes in the room’s acoustic properties, as characterized by the RIR. Existing datasets of RIRs either do not systematically vary positions of objects in an environment, or they consist of only simulated RIRs. We present SoundCam, the largest dataset of unique RIRs from in-the-wild rooms publicly released to date. It includes 5,000 10-channel real-world measurements of room impulse responses and 2,000 10-channel recordings of music in three different rooms, including a controlled acoustic lab, an in-the-wild living room, and a conference room, with different humans in positions throughout each room. We show that these measurements can be used for interesting tasks, such as detecting and identifying humans, and tracking their positions. Mason L. Wang, Samuel Clarke, Jui-Hsien Wang, Ruohan Gao, Jiajun Wu 0001	NeurIPS	5
2023	Inferring Hybrid Neural Fluid Fields from Videos abstract We study recovering fluid density and velocity from sparse multiview videos. Existing neural dynamic reconstruction methods predominantly rely on optical flows; therefore, they cannot accurately estimate the density and uncover the underlying velocity due to the inherent visual ambiguities of fluid velocity, as fluids are often shapeless and lack stable visual features. The challenge is further pronounced by the turbulent nature of fluid flows, which calls for properly designed fluid velocity representations. To address these challenges, we propose hybrid neural fluid fields (HyFluid), a neural approach to jointly infer fluid density and velocity fields. Specifically, to deal with visual ambiguities of fluid velocity, we introduce a set of physics-based losses that enforce inferring a physically plausible velocity field, which is divergence-free and drives the transport of density. To deal with the turbulent nature of fluid velocity, we design a hybrid neural velocity representation that includes a base neural velocity field that captures most irrotational energy and a vortex particle-based velocity that models residual turbulent velocity. We show that our method enables recovering vortical flow details. Our approach opens up possibilities for various learning and reconstruction applications centered around 3D incompressible flow, including fluid re-simulation and editing, future prediction, and neural dynamic scene composition. Project website: https://kovenyu.com/HyFluid/ Hong-Xing Yu, Yitong Deng, Bo Zhu 0002, Jiajun Wu 0001	NeurIPS	6
2023	Physically Plausible Animation of Human Upper Body from a Single Image abstract We present a new method for generating controllable, dynamically responsive, and photorealistic human animations. Given an image of a person, our system allows the user to generate Physically plausible Upper Body Animation (PUBA) using interaction in the image space, such as dragging their hand to various locations. We formulate a reinforcement learning problem to train a dynamic model that predicts the person’s next 2D state (i.e., keypoints on the image) conditioned on a 3D action (i.e., joint torque), and a policy that outputs optimal actions to control the person to achieve desired goals. The dynamic model leverages the expressiveness of 3D simulation and the visual realism of 2D videos. PUBA generates 2D keypoint sequences that achieve task goals while being responsive to forceful perturbation. The sequences of keypoints are then translated by a pose-to-image generator to produce the final photorealistic video. Zhengping Zhou, Yung-Yu Chuang, Jiajun Wu 0001, C. Karen Liu	WACV	4
2023	Neurosymbolic Models for Computer Graphics abstract Abstract Procedural models (i.e. symbolic programs that output visual data) are a historically‐popular method for representing graphics content: vegetation, buildings, textures, etc. They offer many advantages: interpretable design parameters, stochastic variations, high‐quality outputs, compact representation, and more. But they also have some limitations, such as the difficulty of authoring a procedural model from scratch. More recently, AI‐based methods, and especially neural networks, have become popular for creating graphic content. These techniques allow users to directly specify desired properties of the artifact they want to create (via examples, constraints, or objectives), while a search, optimization, or learning algorithm takes care of the details. However, this ease of use comes at a cost, as it's often hard to interpret or manipulate these representations. In this state‐of‐the‐art report, we summarize research on neurosymbolic models in computer graphics: methods that combine the strengths of both AI and symbolic programs to represent, generate, and manipulate visual data. We survey recent work applying these techniques to represent 2D shapes, 3D shapes, and materials & textures. Along the way, we situate each prior work in a unified design space for neurosymbolic models, which helps reveal underexplored areas and opportunities for future research. Daniel Ritchie 0001, Paul Guerrero 0001, R. Kenny Jones, Niloy J. Mitra, Adriana Schulz, Karl D. D. Willis, Jiajun Wu 0001	Comput. Graph. Forum	7
2023	Fluid Simulation on Neural Flow Maps abstract We introduce Neural Flow Maps, a novel simulation method bridging the emerging paradigm of implicit neural representations with fluid simulation based on the theory of flow maps, to achieve state-of-the-art simulation of in-viscid fluid phenomena. We devise a novel hybrid neural field representation, Spatially Sparse Neural Fields (SSNF), which fuses small neural networks with a pyramid of overlapping, multi-resolution, and spatially sparse grids, to compactly represent long-term spatiotemporal velocity fields at high accuracy. With this neural velocity buffer in hand, we compute long-term, bidirectional flow maps and their Jacobians in a mechanistically symmetric manner, to facilitate drastic accuracy improvement over existing solutions. These long-range, bidirectional flow maps enable high advection accuracy with low dissipation, which in turn facilitates high-fidelity incompressible flow simulations that manifest intricate vortical structures. We demonstrate the efficacy of our neural fluid simulation in a variety of challenging simulation scenarios, including leapfrogging vortices, colliding vortices, vortex reconnections, as well as vortex generation from moving obstacles and density differences. Our examples show increased performance over existing methods in terms of energy conservation, visual complexity, adherence to experimental observations, and preservation of detailed vortical structures. Yitong Deng, Hong-Xing Yu, Diyang Zhang, Jiajun Wu 0001, Bo Zhu 0002	ACM Trans. Graph.	4
2023	Object Motion Guided Human Motion Synthesis abstract Modeling human behaviors in contextual environments has a wide range of applications in character animation, embodied AI, VR/AR, and robotics. In real-world scenarios, humans frequently interact with the environment and manipulate various objects to complete daily tasks. In this work, we study the problem of full-body human motion synthesis for the manipulation of large-sized objects. We propose Object MOtion guided human MOtion synthesis (OMOMO), a conditional diffusion framework that can generate full-body manipulation behaviors from only the object motion. Since naively applying diffusion models fails to precisely enforce contact constraints between the hands and the object, OMOMO learns two separate denoising processes to first predict hand positions from object motion and subsequently synthesize full-body poses based on the predicted hand positions. By employing the hand positions as an intermediate representation between the two denoising processes, we can explicitly enforce contact constraints, resulting in more physically plausible manipulation motions. With the learned model, we develop a novel system that captures full-body human manipulation motions by simply attaching a smartphone to the object being manipulated. Through extensive experiments, we demonstrate the effectiveness of our proposed pipeline and its ability to generalize to unseen objects. Additionally, as high-quality human-object interaction datasets are scarce, we collect a large-scale dataset consisting of 3D object geometry, object motion, and human motion. Our dataset contains human-object interaction motion for 15 objects, with a total duration of approximately 10 hours. Jiaman Li, Jiajun Wu 0001, C. Karen Liu	ACM Trans. Graph.	2
2023	Editing Motion Graphics Video via Motion Vectorization and Transformation abstract Motion graphics videos are widely used in Web design, digital advertising, animated logos and film title sequences, to capture a viewer's attention. But editing such video is challenging because the video provides a low-level sequence of pixels and frames rather than higher-level structure such as the objects in the video with their corresponding motions and occlusions. We present a motion vectorization pipeline for converting motion graphics video into an SVG motion program that provides such structure. The resulting SVG program can be rendered using any SVG renderer (e.g. most Web browsers) and edited using any SVG editor. We also introduce a program transformation API that facilitates editing of a SVG motion program to create variations that adjust the timing, motions and/or appearances of objects. We show how the API can be used to create a variety of effects including retiming object motion to match a music beat, adding motion textures to objects, and collision preserving appearance changes. Sharon Zhang, Jiaju Ma, Jiajun Wu 0001, Daniel Ritchie 0001, Maneesh Agrawala	ACM Trans. Graph.	3
2022	Revisiting the "Video" in Video-Language Understanding abstract What makes a video task uniquely suited for videos, beyond what can be understood from a single image? Building on recent progress in self-supervised image-language models, we revisit this question in the context of video and language tasks. We propose the atemporal probe (ATP), a new model for video-language analysis which provides a stronger bound on the baseline accuracy of multimodal models constrained by image-level understanding. By applying this model to standard discriminative video and language tasks, such as video question answering and text-to-video retrieval, we characterize the limitations and potential of current video-language benchmarks. We find that understanding of event temporality is often not necessary to achieve strong or state-of-the-art performance, even compared with recent large-scale video-language models and in contexts intended to benchmark deeper video-level understanding. We also demonstrate how ATP can improve both video-language dataset and model design. We describe a technique for leveraging ATP to better disentangle dataset subsets with a higher concentration of temporally challenging data, improving benchmarking efficacy for causal and temporal understanding. Further, we show that effectively integrating ATP into full video-level temporal models can improve efficiency and state-of-the-art accuracy.11Project website: https://stanfordvl.github.io/atp-revisit-video-lang/ Shyamal Buch, Cristóbal Eyzaguirre, Adrien Gaidon, Jiajun Wu 0001, Li Fei-Fei 0001, Juan Carlos Niebles	CVPR	4
2022	ObjectFolder 2.0: A Multisensory Object Dataset for Sim2Real Transfer abstract Objects play a crucial role in our everyday activities. Though multisensory object-centric learning has shown great potential lately, the modeling of objects in prior work is rather unrealistic. ObjectFolder 1.0 is a recent dataset that introduces 100 virtualized objects with visual, acoustic, and tactile sensory data. However, the dataset is small in scale and the multisensory data is of limited quality, hampering generalization to real-world scenarios. We present ObjectFolder 2.0, a large-scale, multisensory dataset of common household objects in the form of implicit neural representations that significantly enhances ObjectFolder 1.0 in three aspects. First, our dataset is 10 times larger in the amount of objects and orders of magnitude faster in rendering time. Second, we significantly improve the multisensory rendering quality for all three modalities. Third, we show that models learned from virtual objects in our dataset successfully transfer to their real-world counterparts in three challenging tasks: object scale estimation, contact localization, and shape reconstruction. ObjectFolder 2.0 offers a new path and testbed for multisensory learning in computer vision and robotics. The dataset is available at https://github.com/rhgao/ObjectFolder. Ruohan Gao, Zilin Si, Yen-Yu Chang, Samuel Clarke, Jeannette Bohg, Li Fei-Fei 0001, Wenzhen Yuan 0001, Jiajun Wu 0001	CVPR	8
2022	Programmatic Concept Learning for Human Motion Description and Synthesis abstract We introduce Programmatic Motion Concepts, a hierarchical motion representation for human actions that captures both low-level motion and high-level description as motion concepts. This representation enables human motion description, interactive editing, and controlled synthesis of novel video sequences within a single framework. We present an architecture that learns this concept representation from paired video and action sequences in a semi-supervised manner. The compactness of our representation also allows us to present a low-resource training recipe for data-efficient learning. By outperforming established baselines, especially in the small data regime, we demonstrate the efficiency and effectiveness of our framework for multiple applications. Sumith Kulal, Jiayuan Mao, Alex Aiken, Jiajun Wu 0001	CVPR	4
2022	Rotationally Equivariant 3D Object Detection abstract Rotation equivariance has recently become a strongly desired property in the 3D deep learning community. Yet most existing methods focus on equivariance regarding a global input rotation while ignoring the fact that rotation symmetry has its own spatial support. Specifically, we consider the object detection problem in 3D scenes, where an object bounding box should be equivariant regarding the object pose, independent of the scene motion. This suggests a new desired property we call object-level rotation equivariance. To incorporate object-level rotation equivariance into 3D object detectors, we need a mechanism to extract equivariant features with local object-level spatial support while being able to model cross-object context information. To this end, we propose Equivariant Object detection Network (EON) with a rotation equivariance suspension design to achieve object-level equivariance. EON can be applied to modern point cloud object detectors, such as VoteNet and PointR-CNN, enabling them to exploit object rotation symmetry in scene-scale inputs. Our experiments on both indoor scene and autonomous driving datasets show that significant improvements are obtained by plugging our EON design into existing state-of-the-art 3D object detectors. Project website: https://kovenyu.com/EON/ Hong-Xing Yu, Jiajun Wu 0001, Li Yi 0001	CVPR	2
2022	Unsupervised Segmentation in Real-World Images via Spelke Object Inference Rahul M. V., Yoni Friedman, Jiajun Wu 0001, Josh Tenenbaum, Dan Yamins, Daniel Bear	ECCV (29)	4
2022	Translating a Visual LEGO Manual to a Machine-Executable Plan Ruocheng Wang, Jiayuan Mao, Chin-Yi Cheng, Jiajun Wu 0001	ECCV (37)	5
2022	Video Extrapolation in Space and Time Jiajun Wu 0001	ECCV (16)	2
2022	Vision-Based Manipulators Need to Also See from Their Hands Kyle Hsu, Moo Jin Kim, Rafael Rafailov, Jiajun Wu 0001, Chelsea Finn	ICLR	4
2022	SDEdit: Guided Image Synthesis and Editing with Stochastic Differential Equations Chenlin Meng, Yang Song 0011, Jiaming Song, Jiajun Wu 0001, Jun-Yan Zhu, Stefano Ermon	ICLR	5
2022	Unsupervised Discovery of Object Radiance Fields Hong-Xing Yu, Leonidas J. Guibas, Jiajun Wu 0001	ICLR	3
2022	E-MAPP: Efficient Multi-Agent Reinforcement Learning with Parallel Program Guidance abstract A critical challenge in multi-agent reinforcement learning(MARL) is for multiple agents to efficiently accomplish complex, long-horizon tasks. The agents often have difficulties in cooperating on common goals, dividing complex tasks, and planning through several stages to make progress. We propose to address these challenges by guiding agents with programs designed for parallelization, since programs as a representation contain rich structural and semantic information, and are widely used as abstractions for long-horizon tasks. Specifically, we introduce Efficient Multi-Agent Reinforcement Learning with Parallel Program Guidance(E-MAPP), a novel framework that leverages parallel programs to guide multiple agents to efficiently accomplish goals that require planning over $10+$ stages. E-MAPP integrates the structural information from a parallel program, promotes the cooperative behaviors grounded in program semantics, and improves the time efficiency via a task allocator. We conduct extensive experiments on a series of challenging, long-horizon cooperative tasks in the Overcooked environment. Results show that E-MAPP outperforms strong baselines in terms of the completion rate, time efficiency, and zero-shot generalization ability by a large margin. Can Chang, Ni Mu, Jiajun Wu 0001, Ling Pan, Huazhe Xu	NeurIPS	3
2022	Unsupervised Learning of Shape Programs with Repeatable Implicit Parts abstract Shape programs encode shape structures by representing object parts as subroutines and constructing the overall shape by composing these subroutines. This usually involves the reuse of subroutines for repeatable parts, enabling the modeling of correlations among shape elements such as geometric similarity. However, existing learning-based shape programs suffer from limited representation capacity, because they use coarse geometry representations such as geometric primitives and low-resolution voxel grids. Further, their training requires manually annotated ground-truth programs, which are expensive to attain. We address these limitations by proposing Shape Programs with Repeatable Implicit Parts (ProGRIP). Using implicit functions to represent parts, ProGRIP greatly boosts the representation capacity of shape programs while preserving the higher-level structure of repetitions and symmetry. Meanwhile, we free ProGRIP from any inaccessible supervised training via devising a matching-based unsupervised training objective. Our empirical studies show that ProGRIP outperforms existing structured representations in both shape reconstruction fidelity and segmentation accuracy of semantic parts. Boyang Deng, Sumith Kulal, Zhengyang Dong, Congyue Deng, Yonglong Tian, Jiajun Wu 0001	NeurIPS	6
2022	Geoclidean: Few-Shot Generalization in Euclidean Geometry abstract Euclidean geometry is among the earliest forms of mathematical thinking. While the geometric primitives underlying its constructions, such as perfect lines and circles, do not often occur in the natural world, humans rarely struggle to perceive and reason with them. Will computer vision models trained on natural images show the same sensitivity to Euclidean geometry? Here we explore these questions by studying few-shot generalization in the universe of Euclidean geometry constructions. We introduce Geoclidean, a domain-specific language for Euclidean geometry, and use it to generate two datasets of geometric concept learning tasks for benchmarking generalization judgements of humans and machines. We find that humans are indeed sensitive to Euclidean geometry and generalize strongly from a few visual examples of a geometric concept. In contrast, low-level and high-level visual features from standard computer vision models pretrained on natural images do not support correct generalization. Thus Geoclidean represents a novel few-shot generalization benchmark for geometric concept learning, where the performance of humans and of AI models diverge. The Geoclidean framework and dataset are publicly available for download. Joy Hsu, Jiajun Wu 0001, Noah D. Goodman	NeurIPS	2
2022	MOMA-LRG: Language-Refined Graphs for Multi-Object Multi-Actor Activity Parsing abstract Video-language models (VLMs), large models pre-trained on numerous but noisy video-text pairs from the internet, have revolutionized activity recognition through their remarkable generalization and open-vocabulary capabilities. While complex human activities are often hierarchical and compositional, most existing tasks for evaluating VLMs focus only on high-level video understanding, making it difficult to accurately assess and interpret the ability of VLMs to understand complex and fine-grained human activities. Inspired by the recently proposed MOMA framework, we define activity graphs as a single universal representation of human activities that encompasses video understanding at the activity, sub-activity, and atomic action level. We redefine activity parsing as the overarching task of activity graph generation, requiring understanding human activities across all three levels. To facilitate the evaluation of models on activity parsing, we introduce MOMA-LRG (Multi-Object Multi-Actor Language-Refined Graphs), a large dataset of complex human activities with activity graph annotations that can be readily transformed into natural language sentences. Lastly, we present a model-agnostic and lightweight approach to adapting and evaluating VLMs by incorporating structured knowledge from activity graphs into VLMs, addressing the individual limitations of language and graphical models. We demonstrate strong performance on few-shot activity parsing, and our framework is intended to foster future research in the joint modeling of videos, graphs, and language. Zelun Luo, Zane Durante, Linden Li, Wanze Xie, Emily Jin, Zhuoyi Huang, Lun Yu Li, Jiajun Wu 0001, Juan Carlos Niebles, Ehsan Adeli-Mosabbeb, Li Fei-Fei 0001	NeurIPS	9
2022	CLEVRER-Humans: Describing Physical and Causal Events the Human Way abstract Building machines that can reason about physical events and their causal relationships is crucial for flexible interaction with the physical world. However, most existing physical and causal reasoning benchmarks are exclusively based on synthetically generated events and synthetic natural language descriptions of the causal relationships. This design brings up two issues. First, there is a lack of diversity in both event types and natural language descriptions; second, causal relationships based on manually-defined heuristics are different from human judgments. To address both shortcomings, we present the CLEVRER-Humans benchmark, a video reasoning dataset for causal judgment of physical events with human labels. We employ two techniques to improve data collection efficiency: first, a novel iterative event cloze task to elicit a new representation of events in videos, which we term Causal Event Graphs (CEGs); second, a data augmentation technique based on neural language generative models. We convert the collected CEGs into questions and answers to be consistent with prior work. Finally, we study a collection of baseline approaches for CLEVRER-Humans question-answering, highlighting great challenges set forth by our benchmark. Jiayuan Mao, Xuelin Yang, Xikun Zhang 0001, Noah D. Goodman, Jiajun Wu 0001	NeurIPS	5
2022	Interaction Modeling with Multiplex Attention abstract Modeling multi-agent systems requires understanding how agents interact. Such systems are often difficult to model because they can involve a variety of types of interactions that layer together to drive rich social behavioral dynamics. Here we introduce a method for accurately modeling multi-agent systems. We present Interaction Modeling with Multiplex Attention (IMMA), a forward prediction model that uses a multiplex latent graph to represent multiple independent types of interactions and attention to account for relations of different strengths. We also introduce Progressive Layer Training, a training strategy for this architecture. We show that our approach outperforms state-of-the-art models in trajectory forecasting and relation inference, spanning three multi-agent scenarios: social navigation, cooperative task achievement, and team sports. We further demonstrate that our approach can improve zero-shot generalization and allows us to probe how different interactions impact agent behavior. Fan-Yun Sun, Isaac Kauvar, Jiachen Li 0001, Mykel J. Kochenderfer, Jiajun Wu 0001, Nick Haber	NeurIPS	6
2022	IKEA-Manual: Seeing Shape Assembly Step by Step abstract Human-designed visual manuals are crucial components in shape assembly activities. They provide step-by-step guidance on how we should move and connect different parts in a convenient and physically-realizable way. While there has been an ongoing effort in building agents that perform assembly tasks, the information in human-design manuals has been largely overlooked. We identify that this is due to 1) a lack of realistic 3D assembly objects that have paired manuals and 2) the difficulty of extracting structured information from purely image-based manuals. Motivated by this observation, we present IKEA-Manual, a dataset consisting of 102 IKEA objects paired with assembly manuals. We provide fine-grained annotations on the IKEA objects and assembly manuals, including decomposed assembly parts, assembly plans, manual segmentation, and 2D-3D correspondence between 3D parts and visual manuals. We illustrate the broad application of our dataset on four tasks related to shape assembly: assembly plan generation, part segmentation, pose estimationand 3D part assembly. Ruocheng Wang, Jiayuan Mao, Chin-Yi Cheng, Jiajun Wu 0001	NeurIPS	6
2022	Scene Synthesis from Human Motion abstract Large-scale capture of human motion with diverse, complex scenes, while immensely useful, is often considered prohibitively costly. Meanwhile, human motion alone contains rich information about the scene they reside in and interact with. For example, a sitting human suggests the existence of a chair, and their leg position further implies the chair’s pose. In this paper, we propose to synthesize diverse, semantically reasonable, and physically plausible scenes based on human motion. Our framework, Scene Synthesis from HUMan MotiON (SUMMON), includes two steps. It first uses ContactFormer, our newly introduced contact predictor, to obtain temporally consistent contact labels from human motion. Based on these predictions, SUMMON then chooses interacting objects and optimizes physical plausibility losses; it further populates the scene with objects that do not interact with humans. Experimental results demonstrate that SUMMON synthesizes feasible, plausible, and diverse scenes and has the potential to generate extensive human-scene interaction data for the community. Sifan Ye, Yixing Wang, Jiaman Li, Dennis Park, C. Karen Liu, Huazhe Xu, Jiajun Wu 0001	SIGGRAPH Asia	7
2021	Augmenting Policy Learning with Routines Discovered from a Single Demonstration abstract Humans can abstract prior knowledge from very little data and use it to boost skill learning. In this paper, we propose routine-augmented policy learning (RAPL), which discovers routines composed of primitive actions from a single demonstration and uses discovered routines to augment policy learning. To discover routines from the demonstration, we first abstract routine candidates by identifying grammar over the demonstrated action trajectory. Then, the best routines measured by length and frequency are selected to form a routine library. We propose to learn policy simultaneously at primitive-level and routine-level with discovered routines, leveraging the temporal structure of routines. Our approach enables imitating expert behavior at multiple temporal scales for imitation learning and promotes reinforcement learning exploration. Extensive experiments on Atari games demonstrate that RAPL improves the state-of-the-art imitation learning method SQIL and reinforcement learning method A2C. Further, we show that discovered routines can generalize to unseen levels and difficulties on the CoinRun benchmark. Zelin Zhao 0001, Chuang Gan 0001, Jiajun Wu 0001, Josh Tenenbaum	AAAI	3
2021	Pi-GAN: Periodic Implicit Generative Adversarial Networks for 3D-Aware Image Synthesis abstract We have witnessed rapid progress on 3D-aware image synthesis, leveraging recent advances in generative visual models and neural rendering. Existing approaches how-ever fall short in two ways: first, they may lack an under-lying 3D representation or rely on view-inconsistent rendering, hence synthesizing images that are not multi-view consistent; second, they often depend upon representation network architectures that are not expressive enough, and their results thus lack in image quality. We propose a novel generative model, named Periodic Implicit Generative Adversarial Networks (π-GAN or pi-GAN), for high-quality 3D-aware image synthesis. π-GAN leverages neural representations with periodic activation functions and volumetric rendering to represent scenes as view-consistent radiance fields. The proposed approach obtains state-of-the-art results for 3D-aware image synthesis with multiple real and synthetic datasets. Eric R. Chan, Marco Monteiro, Petr Kellnhofer, Jiajun Wu 0001, Gordon Wetzstein	CVPR	4
2021	KeypointDeformer: Unsupervised 3D Keypoint Discovery for Shape Control abstract We introduce KeypointDeformer, a novel unsupervised method for shape control through automatically discovered 3D keypoints. We cast this as the problem of aligning a source 3D object to a target 3D object from the same object category. Our method analyzes the difference between the shapes of the two objects by comparing their latent representations. This latent representation is in the form of 3D keypoints that are learned in an unsupervised way. The difference between the 3D keypoints of the source and the target objects then informs the shape deformation algorithm that deforms the source object into the target object. The whole model is learned end-to-end and simultaneously discovers 3D keypoints while learning to use them for deforming object shapes. Our approach produces intuitive and semantically consistent control of shape deformations. Moreover, our discovered 3D keypoints are consistent across object category instances despite large shape variations. As our method is unsupervised, it can be readily deployed to new object categories without requiring annotations for 3D keypoints and deformations. Project page: http://tomasjakab.github.io/KeypointDeformer. Tomas Jakab, Richard Tucker 0001, Ameesh Makadia, Jiajun Wu 0001, Noah Snavely, Angjoo Kanazawa	CVPR	4
2021	Hierarchical Motion Understanding via Motion Programs abstract Current approaches to video analysis of human motion focus on raw pixels or keypoints as the basic units of reasoning. We posit that adding higher-level motion primitives, which can capture natural coarser units of motion such as "backswing" or "follow-through", can be used to improve downstream analysis tasks. This higher level of abstraction can also capture key features, such as loops of repeated primitives, that are currently inaccessible at lower levels of representation. We therefore introduce Motion Programs, a neuro-symbolic, program-like representation that expresses motions as a composition of high-level primitives. We also present a system for automatically inducing motion programs from videos of human motion and for leveraging motion programs in video synthesis. Experiments show that motion programs can accurately describe a diverse set of human motions and the inferred programs contain semantically meaningful motion primitives, such as arm swings and jumping jacks. Our representation also benefits downstream tasks such as video interpolation and video prediction and outperforms off-the-shelf models. We further demonstrate how these programs can detect diverse kinds of repetitive motion and facilitate interactive video editing. Sumith Kulal, Jiayuan Mao, Alex Aiken, Jiajun Wu 0001	CVPR	4
2021	Repopulating Street Scenes abstract We present a framework for automatically reconfiguring images of street scenes by populating, depopulating, or repopulating them with objects such as pedestrians or vehicles. Applications of this method include anonymizing images to enhance privacy, generating data augmentations for perception tasks like autonomous driving, and composing scenes to achieve a certain ambiance, such as empty streets in the early morning. At a technical level, our work has three primary contributions: (1) a method for clearing images of objects, (2) a method for estimating sun direction from a single image, and (3) a way to compose objects in scenes that respects scene geometry and illumination. Each component is learned from data with minimal ground truth annotations, by making creative use of large-numbers of short image bursts of street scenes. We demonstrate convincing results on a range of street scenes and illustrate potential applications. Yifan Wang 0013, Andrew Liu 0001, Richard Tucker 0001, Jiajun Wu 0001, Brian Curless, Steven M. Seitz, Noah Snavely	CVPR	4
2021	De-Rendering the World's Revolutionary Artefacts abstract Recent works have shown exciting results in unsupervised image de-rendering—learning to decompose 3D shape, appearance, and lighting from single-image collections without explicit supervision. However, many of these assume simplistic material and lighting models. We propose a method, termed RADAR, that can recover environment illumination and surface materials from real single-image collections, relying neither on explicit 3D supervision, nor on multi-view or multi-light images. Specifically, we focus on rotationally symmetric artefacts that exhibit challenging surface properties including specular reflections, such as vases. We introduce a novel self-supervised albedo discriminator, which allows the model to recover plausible albedo without requiring any ground-truth during training. In conjunction with a shape reconstruction module exploiting rotational symmetry, we present an end-to-end learning framework that is able to de-render the world’s revolutionary artefacts. We conduct experiments on a real vase dataset and demonstrate compelling decomposition results, allowing for applications including free-viewpoint rendering and relighting. More results and code at: https://sorderender.github.io/. Shangzhe Wu, Ameesh Makadia, Jiajun Wu 0001, Noah Snavely, Richard Tucker 0001, Angjoo Kanazawa	CVPR	3
2021	Neural Radiance Flow for 4D View Synthesis and Video Processing abstract We present a method, Neural Radiance Flow (NeRFlow), to learn a 4D spatial-temporal representation of a dynamic scene from a set of RGB images. Key to our approach is the use of a neural implicit representation that learns to capture the 3D occupancy, radiance, and dynamics of the scene. By enforcing consistency across different modalities, our representation enables multi-view rendering in diverse dynamic scenes, including water pouring, robotic interaction, and real images, outperforming state-of-the-art methods for spatial-temporal view synthesis. Our approach works even when being provided only a single monocular real video. We further demonstrate that the learned representation can serve as an implicit scene prior, enabling video processing tasks such as image super-resolution and de-noising without any additional supervision. Yilun Du, Hong-Xing Yu, Josh Tenenbaum, Jiajun Wu 0001	ICCV	5
2021	Learning Temporal Dynamics from Cycles in Narrated Video abstract Learning to model how the world changes as time elapses has proven a challenging problem for the computer vision community. We introduce a self-supervised approach to this problem that solves a multi-modal temporal cycle consistency objective jointly in vision and language. This objective requires a model to learn modality-agnostic functions to predict the future and past that undo each other when composed. We hypothesize that a model trained on this objective will discover long-term temporal dynamics in video. We verify this hypothesis by using the resultant visual representations and predictive models as-is to solve a variety of downstream tasks. Our method outperforms state-of-the-art self-supervised video prediction methods on future action anticipation, temporal image ordering, and arrow-of-time classification tasks, without training on target datasets or their labels. Dave Epstein, Jiajun Wu 0001, Cordelia Schmid, Chen Sun 0002	ICCV	2
2021	3D Shape Generation and Completion through Point-Voxel Diffusion abstract We propose a novel approach for probabilistic generative modeling of 3D shapes. Unlike most existing models that learn to deterministically translate a latent vector to a shape, our model, Point-Voxel Diffusion (PVD), is a unified, probabilistic formulation for unconditional shape generation and conditional, multi-modal shape completion. PVD marries denoising diffusion models with the hybrid, point-voxel representation of 3D shapes. It can be viewed as a series of denoising steps, reversing the diffusion process from observed point cloud data to Gaussian noise, and is trained by optimizing a variational lower bound to the (conditional) likelihood function. Experiments demonstrate that PVD is capable of synthesizing high-fidelity shapes, completing partial point clouds, and generating multiple completion results from single-view depth scans of real objects. Linqi Zhou, Yilun Du, Jiajun Wu 0001	ICCV	3
2021	Grounding Physical Concepts of Objects and Events Through Dynamic Visual Reasoning Zhenfang Chen, Jiayuan Mao, Jiajun Wu 0001, Kwan-Yee Kenneth Wong, Josh Tenenbaum, Chuang Gan 0001	ICLR	3
2021	Unsupervised Discovery of 3D Physical Objects from Video Yilun Du, Kevin A. Smith 0001, Tomer D. Ullman, Josh Tenenbaum, Jiajun Wu 0001	ICLR	5
2021	Temporal and Object Quantification Networks abstract We present Temporal and Object Quantification Networks (TOQ-Nets), a new class of neuro-symbolic networks with a structural bias that enables them to learn to recognize complex relational-temporal events. This is done by including reasoning layers that implement finite-domain quantification over objects and time. The structure allows them to generalize directly to input instances with varying numbers of objects in temporal sequences of varying lengths. We evaluate TOQ-Nets on input domains that require recognizing event-types in terms of complex temporal relational patterns. We demonstrate that TOQ-Nets can generalize from small amounts of data to scenarios containing more objects than were present during training and to temporal warpings of input sequences. Jiayuan Mao, Zhezheng Luo, Chuang Gan 0001, Josh Tenenbaum, Jiajun Wu 0001, Leslie Pack Kaelbling, Tomer D. Ullman	IJCAI	5
2021	Grammar-Based Grounded Lexicon Learning abstract We present Grammar-Based Grounded Language Learning (G2L2), a lexicalist approach toward learning a compositional and grounded meaning representation of language from grounded data, such as paired images and texts. At the core of G2L2 is a collection of lexicon entries, which map each word to a tuple of a syntactic type and a neuro-symbolic semantic program. For example, the word shiny has a syntactic type of adjective; its neuro-symbolic semantic program has the symbolic form $\lambda x.\textit{filter}(x, \textbf{SHINY})$, where the concept SHINY is associated with a neural network embedding, which will be used to classify shiny objects. Given an input sentence, G2L2 first looks up the lexicon entries associated with each token. It then derives the meaning of the sentence as an executable neuro-symbolic program by composing lexical meanings based on syntax. The recovered meaning programs can be executed on grounded inputs. To facilitate learning in an exponentially-growing compositional space, we introduce a joint parsing and expected execution algorithm, which does local marginalization over derivations to reduce the training time. We evaluate G2L2 on two domains: visual reasoning and language-driven navigation. Results show that G2L2 can generalize from small amounts of data to novel compositions of words. Jiayuan Mao, Freda Shi, Jiajun Wu 0001, Roger Levy, Josh Tenenbaum	NeurIPS	3
2021	DASH: Modularized Human Manipulation Simulation with Vision and Language for Embodied AI abstract Creating virtual humans with embodied, human-like perceptual and actuation constraints has the promise to provide an integrated simulation platform for many scientific and engineering applications. We present Dynamic and Autonomous Simulated Human (DASH), an embodied virtual human that, given natural language commands, performs grasp-and-stack tasks in a physically-simulated cluttered environment solely using its own visual perception, proprioception, and touch, without requiring human motion data. By factoring the DASH system into a vision module, a language module, and manipulation modules of two skill categories, we can mix and match analytical and machine learning techniques for different modules so that DASH is able to not only perform randomly arranged tasks with a high success rate, but also do so under anthropomorphic constraints and with fluid and diverse motions. The modular design also favors analysis and extensibility to more complex manipulation skills. Yifeng Jiang 0002, Michelle Guo, Jiangshan Li, Ioannis Exarchos, Jiajun Wu 0001, C. Karen Liu	SCA	5
2021	When is particle filtering efficient for planning in partially observed linear dynamical systems? abstract Particle filtering is a popular method for inferring latent states in stochastic dynamical systems, whose theoretical properties have been well studied in machine learning and statistics communities. In many control problems, e.g., partially observed linear dynamical systems (POLDS), oftentimes the inferred latent state is further used for planning at each step. This paper initiates a rigorous study on the efficiency of particle filtering for sequential planning, and gives the first particle complexity bounds. Though errors in past actions may affect the future, we are able to bound the number of particles needed so that the long-run reward of the policy based on particle filtering is close to that based on exact inference. In particular, we show that, in stable systems, polynomially many particles suffice. Key in our proof is a coupling of the ideal sequence based on the exact planning and the sequence generated by approximate planning based on particle filtering. We believe this technique can be useful in other sequential decision-making problems. Simon S. Du, Wei Hu 0014, Zhiyuan Li 0005, Ruoqi Shen, Zhao Song 0002, Jiajun Wu 0001	UAI	6
2020	The fine structure of surprise in intuitive physics: when, why, and how much? Kevin A. Smith 0001, Lingjie Mei, Shunyu Yao 0006, Jiajun Wu 0001, Elizabeth S. Spelke, Josh Tenenbaum, Tomer D. Ullman	CogSci	4
2020	Perspective Plane Program Induction From a Single Image abstract We study the inverse graphics problem of inferring a holistic representation for natural images. Given an input image, our goal is to induce a neuro-symbolic, program-like representation that jointly models camera poses, object locations, and global scene structures. Such high-level, holistic scene representations further facilitate low-level image manipulation tasks such as inpainting. We formulate this problem as jointly finding the camera pose and scene structure that best describe the input image. The benefits of such joint inference are two-fold: scene regularity serves as a new cue for perspective correction, and in turn, correct perspective correction leads to a simplified scene structure, similar to how the correct shape leads to the most regular texture in shape from texture. Our proposed framework, Perspective Plane Program Induction (P3I), combines search-based and gradient-based algorithms to efficiently solve the problem. P3I outperforms a set of baselines on a collection of Internet images, across tasks including camera pose estimation, global structure inference, and down-stream image manipulation tasks. Jiayuan Mao, Xiuming Zhang, William T. Freeman, Josh Tenenbaum, Jiajun Wu 0001	CVPR	6
2020	End-to-End Optimization of Scene Layout abstract We propose an end-to-end variational generative model for scene layout synthesis conditioned on scene graphs. Unlike unconditional scene layout generation, we use scene graphs as an abstract but general representation to guide the synthesis of diverse scene layouts that satisfy relationships included in the scene graph. This gives rise to more flexible control over the synthesis process, allowing various forms of inputs such as scene layouts extracted from sentences or inferred from a single color image. Using our conditional layout synthesizer, we can generate various layouts that share the same structure of the input example. In addition to this conditional generation design, we also integrate a differentiable rendering module that enables layout refinement using only 2D projections of the scene. Given a depth and a semantics map, the differentiable rendering module enables optimizing over the synthesized layout to fit the given input in an analysis-by-synthesis fashion. Experiments suggest that our model achieves higher accuracy and diversity in conditional scene synthesis and allows exemplar-based scene generation from various input forms. Andrew Luo 0001, Zhoutong Zhang, Jiajun Wu 0001, Josh Tenenbaum	CVPR	3
2020	Probabilistic Video Prediction From Noisy Data With a Posterior Confidence abstract We study a new research problem of probabilistic future frames prediction from a sequence of noisy inputs, which is useful because it is difficult to guarantee the quality of input frames in practical spatiotemporal prediction applications. It is also challenging because it involves two levels of uncertainty: the perceptual uncertainty from noisy observations and the dynamics uncertainty in forward modeling. In this paper, we propose to tackle this problem with an end-to-end trainable model named Bayesian Predictive Network (BP-Net). Unlike previous work in stochastic video prediction that assumes spatiotemporal coherence and therefore fails to deal with perceptual uncertainty, BP-Net models both levels of uncertainty in an integrated framework. Furthermore, unlike previous work that can only provide unsorted estimations of future frames, BP-Net leverages a differentiable sequential importance sampling (SIS) approach to make future predictions based on the inference of underlying physical states, thereby providing sorted prediction candidates in accordance with the SIS importance weights, i.e., the confidences. Our experiment results demonstrate that BP-Net remarkably outperforms existing approaches on predicting future frames from noisy data. Yunbo Wang, Jiajun Wu 0001, Mingsheng Long, Josh Tenenbaum	CVPR	2
2020	Learning Compositional Koopman Operators for Model-Based Control Yunzhu Li, Hao He 0011, Jiajun Wu 0001, Dina Katabi, Antonio Torralba 0001	ICLR	3
2020	CLEVRER: Collision Events for Video Representation and Reasoning Kexin Yi, Chuang Gan 0001, Yunzhu Li, Pushmeet Kohli, Jiajun Wu 0001, Antonio Torralba 0001, Josh Tenenbaum	ICLR	5
2020	Deep Audio Priors Emerge From Harmonic Convolutional Networks Zhoutong Zhang, Chuang Gan 0001, Jiajun Wu 0001, Josh Tenenbaum, Antonio Torralba 0001, William T. Freeman	ICLR	4
2020	Visual Grounding of Learned Physical Models abstract Humans intuitively recognize objects’ physical properties and predict their motion, even when the objects are engaged in complicated interactions. The abilities to perform physical reasoning and to adapt to new environments, while intrinsic to humans, remain challenging to state-of-the-art computational models. In this work, we present a neural model that simultaneously reasons about physics and makes future predictions based on visual and dynamics priors. The visual prior predicts a particle-based representation of the system from visual observations. An inference module operates on those particles, predicting and refining estimates of particle locations, object states, and physical parameters, subject to the constraints imposed by the dynamics prior, which we refer to as visual grounding. We demonstrate the effectiveness of our method in environments involving rigid objects, deformable materials, and fluids. Experiments show that our model can infer the physical properties within a few observations, which allows the model to quickly adapt to unseen scenarios and make accurate predictions into the future. Yunzhu Li, Toru Lin, Kexin Yi, Daniel Bear, Dan Yamins, Jiajun Wu 0001, Josh Tenenbaum, Antonio Torralba 0001	ICML	6
2020	Look, Listen, and Act: Towards Audio-Visual Embodied Navigation abstract A crucial ability of mobile intelligent agents is to integrate the evidence from multiple sensory inputs in an environment and to make a sequence of actions to reach their goals. In this paper, we attempt to approach the problem of Audio-Visual Embodied Navigation, the task of planning the shortest path from a random starting location in a scene to the sound source in an indoor environment, given only raw egocentric visual and audio sensory data. To accomplish this task, the agent is required to learn from various modalities, i.e., relating the audio signal to the visual environment. Here we describe an approach to audio-visual embodied navigation that takes advantage of both visual and audio pieces of evidence. Our solution is based on three key ideas: a visual perception mapper module that constructs its spatial memory of the environment, a sound perception module that infers the relative location of the sound source from the agent, and a dynamic path planner that plans a sequence of actions based on the audio-visual observations and the spatial memory of the environment to navigate toward the goal. Experimental results on a newly collected Visual-Audio-Room dataset using the simulated multi-modal environment demonstrate the effectiveness of our approach over several competitive baselines. Chuang Gan 0001, Yiwei Zhang 0010, Jiajun Wu 0001, Boqing Gong, Josh Tenenbaum	ICRA	3
2020	Accurate Vision-based Manipulation through Contact Reasoning abstract Planning contact interactions is one of the core challenges of many robotic tasks. Optimizing contact locations while taking dynamics into account is computationally costly and, in environments that are only partially observable, executing contact-based tasks often suffers from low accuracy. We present an approach that addresses these two challenges for the problem of vision-based manipulation. First, we propose to disentangle contact from motion optimization. Thereby, we improve planning efficiency by focusing computation on promising contact locations. Second, we use a hybrid approach for perception and state estimation that combines neural networks with a physically meaningful state representation. In simulation and real-world experiments on the task of planar pushing, we show that our method is more efficient and achieves a higher manipulation accuracy than previous vision-based approaches. Alina Kloss, Maria Bauzá 0001, Jiajun Wu 0001, Josh Tenenbaum, Alberto Rodriguez 0003, Jeannette Bohg	ICRA	3
2020	DualSMC: Tunneling Differentiable Filtering and Planning under Continuous POMDPs abstract A major difficulty of solving continuous POMDPs is to infer the multi-modal distribution of the unobserved true states and to make the planning algorithm dependent on the perceived uncertainty. We cast POMDP filtering and planning problems as two closely related Sequential Monte Carlo (SMC) processes, one over the real states and the other over the future optimal trajectories, and combine the merits of these two parts in a new model named the DualSMC network. In particular, we first introduce an adversarial particle filter that leverages the adversarial relationship between its internal components. Based on the filtering results, we then propose a planning algorithm that extends the previous SMC planning approach [Piche et al., 2018] to continuous POMDPs with an uncertainty-dependent policy. Crucially, not only can DualSMC handle complex observations such as image input but also it remains highly interpretable. It is shown to be effective in three continuous POMDP domains: the floor positioning domain, the 3D light-dark navigation domain, and a modified Reacher domain. Yunbo Wang, Bo Liu 0042, Jiajun Wu 0001, Yuke Zhu, Simon S. Du, Li Fei-Fei 0001, Josh Tenenbaum	IJCAI	3
2020	Learning Physical Graph Representations from Visual Scenes abstract Convolutional Neural Networks (CNNs) have proved exceptional at learning representations for visual object categorization. However, CNNs do not explicitly encode objects, parts, and their physical properties, which has limited CNNs' success on tasks that require structured understanding of visual scenes. To overcome these limitations, we introduce the idea of ``Physical Scene Graphs'' (PSGs), which represent scenes as hierarchical graphs, with nodes in the hierarchy corresponding intuitively to object parts at different scales, and edges to physical connections between parts. Bound to each node is a vector of latent attributes that intuitively represent object properties such as surface shape and texture. We also describe PSGNet, a network architecture that learns to extract PSGs by reconstructing scenes through a PSG-structured bottleneck. PSGNet augments standard CNNs by including: recurrent feedback connections to combine low and high-level image information; graph pooling and vectorization operations that convert spatially-uniform feature maps into object-centric graph structures; and perceptual grouping principles to encourage the identification of meaningful scene elements. We show that PSGNet outperforms alternative self-supervised scene representation algorithms at scene segmentation tasks, especially on complex real-world images, and generalizes well to unseen object types and scene arrangements. PSGNet is also able learn from physical motion, enhancing scene estimates even for static images. We present a series of ablation studies illustrating the importance of each component of the PSGNet architecture, analyses showing that learned latent attributes capture intuitive scene properties, and illustrate the use of PSGs for compositional scene inference. Daniel Bear, Chaofei Fan, Damian Mrowca, Yunzhu Li, Seth Alter, Aran Nayebi, Jeremy Schwartz, Li Fei-Fei 0001, Jiajun Wu 0001, Josh Tenenbaum, Dan Yamins	NeurIPS	9
2020	Multi-Plane Program Induction with 3D Box Priors abstract We consider two important aspects in understanding and editing images: modeling regular, program-like texture or patterns in 2D planes, and 3D posing of these planes in the scene. Unlike prior work on image-based program synthesis, which assumes the image contains a single visible 2D plane, we present Box Program Induction (BPI), which infers a program-like scene representation that simultaneously models repeated structure on multiple 2D planes, the 3D position and orientation of the planes, and camera parameters, all from a single image. Our model assumes a box prior, i.e., that the image captures either an inner view or an outer view of a box in 3D. It uses neural networks to infer visual cues such as vanishing points, wireframe lines to guide a search-based algorithm to find the program that best explains the image. Such a holistic, structured scene representation enables 3D-aware interactive image editing operations such as inpainting missing pixels, changing camera parameters, and extrapolate the image contents. Jiayuan Mao, Xiuming Zhang, William T. Freeman, Josh Tenenbaum, Noah Snavely, Jiajun Wu 0001	NeurIPS	7
2020	Learning Generative Models of 3D Structures abstract Abstract 3D models of objects and scenes are critical to many academic disciplines and industrial applications. Of particular interest is the emerging opportunity for 3D graphics to serve artificial intelligence: computer vision systems can benefit from synthetically‐generated training data rendered from virtual 3D scenes, and robots can be trained to navigate in and interact with real‐world environments by first acquiring skills in simulated ones. One of the most promising ways to achieve this is by learning and applyinggenerative modelsof 3D content: computer programs that can synthesize new 3D shapes and scenes. To allow users to edit and manipulate the synthesized 3D content to achieve their goals, the generative model should also bestructure‐aware: it should express 3D shapes and scenes using abstractions that allow manipulation of their high‐level structure. This state‐of‐the‐art report surveys historical work and recent progress on learning structure‐aware generative models of 3D shapes and scenes. We present fundamental representations of 3D shape and scene geometry and structures, describe prominent methodologies including probabilistic models, deep generative models, program synthesis, and neural networks for structured data, and cover many recent methods for structure‐aware synthesis of 3D shapes and indoor scenes. Siddhartha Chaudhuri, Daniel Ritchie 0001, Jiajun Wu 0001, Kai Xu 0004, Hao (Richard) Zhang	Comput. Graph. Forum	3
2019	Real-time inference of physical properties in dynamic scenes Kevin A. Smith 0001, Mario Belledonne, Ilker Yildirim, Jiajun Wu 0001, Josh Tenenbaum	CogSci	4
2019	Program-Guided Image Manipulators abstract Humans are capable of building holistic representations for images at various levels, from local objects, to pairwise relations, to global structures. The interpretation of structures involves reasoning over repetition and symmetry of the objects in the image. In this paper, we present the Program-Guided Image Manipulator (PG-IM), inducing neuro-symbolic program-like representations to represent and manipulate images. Given an image, PG-IM detects repeated patterns, induces symbolic programs, and manipulates the image using a neural network that is guided by the program. PG-IM learns from a single image, exploiting its internal statistics. Despite trained only on image inpainting, PG-IM is directly capable of extrapolation and regularity editing in a unified framework. Extensive experiments show that PG-IM achieves superior performance on all the tasks. Xiuming Zhang, Jiayuan Mao, William T. Freeman, Josh Tenenbaum, Jiajun Wu 0001	ICCV	6
2019	Reasoning About Physical Interactions with Object-Oriented Prediction and Planning Michael Janner, Sergey Levine, William T. Freeman, Josh Tenenbaum, Chelsea Finn, Jiajun Wu 0001	ICLR (Poster)	6
2019	Learning Particle Dynamics for Manipulating Rigid Bodies, Deformable Objects, and Fluids Yunzhu Li, Jiajun Wu 0001, Russ Tedrake, Josh Tenenbaum, Antonio Torralba 0001	ICLR (Poster)	2
2019	Learning to Describe Scenes with Programs Daniel Ritchie 0001, William T. Freeman, Josh Tenenbaum, Jiajun Wu 0001	ICLR (Poster)	6
2019	The Neuro-Symbolic Concept Learner: Interpreting Scenes, Words, and Sentences From Natural Supervision Jiayuan Mao, Chuang Gan 0001, Pushmeet Kohli, Josh Tenenbaum, Jiajun Wu 0001	ICLR	5
2019	Stochastic Prediction of Multi-Agent Interactions from Partial Observations Chen Sun 0002, Per Karlsson, Jiajun Wu 0001, Josh Tenenbaum, Kevin Murphy 0002	ICLR (Poster)	3
2019	Learning to Infer and Execute 3D Shape Programs Yonglong Tian, Andrew Luo 0001, Xingyuan Sun, Kevin Ellis, William T. Freeman, Josh Tenenbaum, Jiajun Wu 0001	ICLR (Poster)	7
2019	Unsupervised Discovery of Parts, Structure, and Dynamics Zhenjia Xu, Chen Sun 0002, Kevin Murphy 0002, William T. Freeman, Josh Tenenbaum, Jiajun Wu 0001	ICLR (Poster)	7
2019	Neurally-Guided Structure Inference abstract Most structure inference methods either rely on exhaustive search or are purely data-driven. Exhaustive search robustly infers the structure of arbitrarily complex data, but it is slow. Data-driven methods allow efficient inference, but do not generalize when test data have more complex structures than training data. In this paper, we propose a hybrid inference algorithm, the Neurally-Guided Structure Inference (NG-SI), keeping the advantages of both search-based and data-driven methods. The key idea of NG-SI is to use a neural network to guide the hierarchical, layer-wise search over the compositional space of structures. We evaluate our algorithm on two representative structure inference tasks: probabilistic matrix decomposition and symbolic program parsing. It outperforms data-driven and search-based alternatives on both tasks. Sidi Lu, Jiayuan Mao, Josh Tenenbaum, Jiajun Wu 0001	ICML	4
2019	Combining Physical Simulators and Object-Based Networks for Control abstract Physics engines play an important role in robot planning and control; however, many real-world control problems involve complex contact dynamics that cannot be characterized analytically. Most physics engines therefore employ approximations that lead to a loss in precision. In this paper, we propose a hybrid dynamics model, simulator-augmented interaction networks (SAIN), combining a physics engine with an object-based neural network for dynamics modeling. Compared with existing models that are purely analytical or purely data-driven, our hybrid model captures the dynamics of interacting objects in a more accurate and data-efficient manner. Experiments both in simulation and on a real robot suggest that it also leads to better performance when used in complex control tasks. Finally, we show that our model generalizes to novel environments with varying object shapes and materials. Anurag Ajay, Maria Bauzá 0001, Jiajun Wu 0001, Nima Fazeli, Josh Tenenbaum, Alberto Rodriguez 0003, Leslie Pack Kaelbling	ICRA	3
2019	ChainQueen: A Real-Time Differentiable Physical Simulator for Soft Robotics abstract Physical simulators have been widely used in robot planning and control. Among them, differentiable simulators are particularly favored, as they can be incorporated into gradient-based optimization algorithms that are efficient in solving inverse problems such as optimal control and motion planning. Therefore, rigid body simulators and recently their differentiable variants are studied extensively. Simulating deformable objects is, however, more challenging compared to rigid body dynamics. The underlying physical laws of deformable objects are more complex, and the resulting systems have orders of magnitude more degrees of freedom and there-fore they are significantly more computationally expensive to simulate. Computing gradients with respect to physical design or controller parameters is typically even more computationally challenging. In this paper, we propose a real-time, differentiable hybrid Lagrangian-Eulerian physical simulator for deformable objects, ChainQueen, based on the Moving Least Squares Material Point Method (MLS-MPM). MLS-MPM can simulate deformable objects with collisions and can be seamlessly incorporated into soft robotic systems. We demonstrate that our simulator achieves high precision in both forward simulation and backward gradient computation. We have successfully employed it in a diverse set of inference, control and co-design tasks for soft robotics. Yuanming Hu, Jiancheng Liu, Andrew Spielberg, Josh Tenenbaum, William T. Freeman, Jiajun Wu 0001, Daniela Rus, Wojciech Matusik	ICRA	6
2019	Propagation Networks for Model-Based Control Under Partial Observation abstract There has been an increasing interest in learning dynamics simulators for model-based control. Compared with off-the-shelf physics engines, a learnable simulator can quickly adapt to unseen objects, scenes, and tasks. However, existing models like interaction networks only work for fully observable systems; they also only consider pairwise interactions within a single time step, both restricting their use in practical systems. We introduce Propagation Networks (PropNet), a differentiable, learnable dynamics model that handles partially observable scenarios and enables instantaneous propagation of signals beyond pairwise interactions. With these innovations, our propagation networks not only outperform current learnable physics engines in forward simulation, but also achieves superior performance on various control tasks. Compared with existing deep reinforcement learning algorithms, model-based control with propagation networks is more accurate, efficient, and generalizable to novel, partially observable scenes and tasks. Yunzhu Li, Jiajun Wu 0001, Jun-Yan Zhu, Josh Tenenbaum, Antonio Torralba 0001, Russ Tedrake	ICRA	2
2019	Visual Concept-Metaconcept Learning abstract Humans reason with concepts and metaconcepts: we recognize red and blue from visual input; we also understand that they are colors, i.e., red is an instance of color. In this paper, we propose the visual concept-metaconcept learner (VCML) for joint learning of concepts and metaconcepts from images and associated question-answer pairs. The key is to exploit the bidirectional connection between visual concepts and metaconcepts. Visual representations provide grounding cues for predicting relations between unseen pairs of concepts. Knowing that red and blue are instances of color, we generalize to the fact that green is also an instance of color since they all categorize the hue of objects. Meanwhile, knowledge about metaconcepts empowers visual concept learning from limited, noisy, and even biased data. From just a few examples of purple cubes we can understand a new color purple, which resembles the hue of the cubes instead of the shape of them. Evaluation on both synthetic and real-world datasets validates our claims. Chi Han, Jiayuan Mao, Chuang Gan 0001, Josh Tenenbaum, Jiajun Wu 0001	NeurIPS	5
2019	Modeling Expectation Violation in Intuitive Physics with Coarse Probabilistic Object Representations abstract From infancy, humans have expectations about how objects will move and interact. Even young children expect objects not to move through one another, teleport, or disappear. They are surprised by mismatches between physical expectations and perceptual observations, even in unfamiliar scenes with completely novel objects. A model that exhibits human-like understanding of physics should be similarly surprised, and adjust its beliefs accordingly. We propose ADEPT, a model that uses a coarse (approximate geometry) object-centric representation for dynamic 3D scene understanding. Inference integrates deep recognition networks, extended probabilistic physical simulation, and particle filtering for forming predictions and expectations across occlusion. We also present a new test set for measuring violations of physical expectations, using a range of scenarios derived from developmental psychology. We systematically compare ADEPT, baseline models, and human expectations on this test set. ADEPT outperforms standard network architectures in discriminating physically implausible scenes, and often performs this discrimination at the same level as people. Kevin A. Smith 0001, Lingjie Mei, Shunyu Yao 0006, Jiajun Wu 0001, Elizabeth S. Spelke, Josh Tenenbaum, Tomer D. Ullman	NeurIPS	4
2019	Video Enhancement with Task-Oriented Flow Tianfan Xue, Baian Chen, Jiajun Wu 0001, Donglai Wei 0001, William T. Freeman	Int. J. Comput. Vis.	3
2019	Visual Dynamics: Stochastic Future Generation via Layered Cross Convolutional Networks abstract We study the problem of synthesizing a number of likely future frames from a single input image. In contrast to traditional methods that have tackled this problem in a deterministic or non-parametric way, we propose to model future frames in a probabilistic manner. Our probabilistic model makes it possible for us to sample and synthesize many possible future frames from a single input image. To synthesize realistic movement of objects, we propose a novel network structure, namely a Cross Convolutional Network; this network encodes image and motion information as feature maps and convolutional kernels, respectively. In experiments, our model performs well on synthetic data, such as 2D shapes and animated game sprites, and on real-world video frames. We present analyses of the learned network representations, showing it is implicitly learning a compact encoding of object appearance and motion. We also demonstrate a few of its applications, including visual analogy-making and video extrapolation. Tianfan Xue, Jiajun Wu 0001, Katherine L. Bouman, William T. Freeman	IEEE Trans. Pattern Anal. Mach. Intell.	2
2018	Attention Clusters: Purely Attention Based Local Feature Integration for Video Classification abstract Recently, substantial research effort has focused on how to apply CNNs or RNNs to better capture temporal patterns in videos, so as to improve the accuracy of video classification. In this paper, however, we show that temporal information, especially longer-term patterns, may not be necessary to achieve competitive results on common trimmed video classification datasets. We investigate the potential of a purely attention based local feature integration. Accounting for the characteristics of such features in video classification, we propose a local feature integration framework based on attention clusters, and introduce a shifting operation to capture more diverse signals. We carefully analyze and compare the effect of different attention mechanisms, cluster sizes, and the use of the shifting operation, and also investigate the combination of attention clusters for multimodal integration. We demonstrate the effectiveness of our framework on three real-world video classification datasets. Our model achieves competitive results across all of these. In particular, on the large-scale Kinetics dataset, our framework obtains an excellent single model accuracy of 79.4% in terms of the top-1 and 94.0% in terms of the top-5 accuracy on the validation set. Xiang Long, Chuang Gan 0001, Gerard de Melo, Jiajun Wu 0001, Xiao Liu 0022, Shilei Wen	CVPR	4
2018	Pix3D: Dataset and Methods for Single-Image 3D Shape Modeling abstract We study 3D shape modeling from a single image and make contributions to it in three aspects. First, we present Pix3D, a large-scale benchmark of diverse image-shape pairs with pixel-level 2D-3D alignment. Pix3D has wide applications in shape-related tasks including reconstruction, retrieval, viewpoint estimation, etc. Building such a large-scale dataset, however, is highly challenging; existing datasets either contain only synthetic data, or lack precise alignment between 2D images and 3D shapes, or only have a small number of images. Second, we calibrate the evaluation criteria for 3D shape reconstruction through behavioral studies, and use them to objectively and systematically benchmark cutting-edge reconstruction algorithms on Pix3D. Third, we design a novel model that simultaneously performs 3D reconstruction and pose estimation; our multi-task learning approach achieves state-of-the-art performance on both tasks. Xingyuan Sun, Jiajun Wu 0001, Xiuming Zhang, Zhoutong Zhang, Chengkai Zhang, Tianfan Xue, Josh Tenenbaum, William T. Freeman	CVPR	2
2018	Physical Primitive Decomposition William T. Freeman, Josh Tenenbaum, Jiajun Wu 0001	ECCV (12)	4
2018	Learning Shape Priors for Single-View 3D Completion And Reconstruction Jiajun Wu 0001, Chengkai Zhang, Xiuming Zhang, Zhoutong Zhang, William T. Freeman, Josh Tenenbaum	ECCV (11)	1
2018	Seeing Tree Structure from Vibration Tianfan Xue, Jiajun Wu 0001, Zhoutong Zhang, Chengkai Zhang, Josh Tenenbaum, William T. Freeman	ECCV (9)	2
2018	Augmenting Physical Simulators with Stochastic Neural Networks: Case Study of Planar Pushing and Bouncing abstract An efficient, generalizable physical simulator with universal uncertainty estimates has wide applications in robot state estimation, planning, and control. In this paper, we build such a simulator for two scenarios, planar pushing and ball bouncing, by augmenting an analytical rigid-body simulator with a neural network that learns to model uncertainty as residuals. Combining symbolic, deterministic simulators with learnable, stochastic neural nets provides us with expressiveness, efficiency, and generalizability simultaneously. Our model outperforms both purely analytical and purely learned simulators consistently on real, standard benchmarks. Compared with methods that model uncertainty using Gaussian processes, our model runs much faster, generalizes better to new object shapes, and is able to characterize the complex distribution of object trajectories. Anurag Ajay, Jiajun Wu 0001, Nima Fazeli, Maria Bauzá 0001, Leslie Pack Kaelbling, Josh Tenenbaum, Alberto Rodriguez 0003	IROS	2
2018	3D Shape Perception from Monocular Vision, Touch, and Shape Priors abstract Perceiving accurate 3D object shape is important for robots to interact with the physical world. Current research along this direction has been primarily relying on visual observations. Vision, however useful, has inherent limitations due to occlusions and the 2D-3D ambiguities, especially for perception with a monocular camera. In contrast, touch gets precise local shape information, though its efficiency for reconstructing the entire shape could be low. In this paper, we propose a novel paradigm that efficiently perceives accurate 3D object shape by incorporating visual and tactile observations, as well as prior knowledge of common object shapes learned from large-scale shape repositories. We use vision first, applying neural networks with learned shape priors to predict an object's 3D shape from a single-view color image. We then use tactile sensing to refine the shape; the robot actively touches the object regions where the visual prediction has high uncertainty. Our method efficiently builds the 3D shape of common objects from a color image and a small number of tactile explorations (around 10). Our setup is easy to apply and has potentials to help robots better perform grasping or manipulation tasks on real-world objects. Shaoxiong Wang, Jiajun Wu 0001, Xingyuan Sun, Wenzhen Yuan 0001, William T. Freeman, Josh Tenenbaum, Edward H. Adelson	IROS	2
2018	Learning to Exploit Stability for 3D Scene Parsing abstract Human scene understanding uses a variety of visual and non-visual cues to perform inference on object types, poses, and relations. Physics is a rich and universal cue which we exploit to enhance scene understanding. We integrate the physical cue of stability into the learning process using a REINFORCE approach coupled to a physics engine, and apply this to the problem of producing the 3D bounding boxes and poses of objects in a scene. We first show that applying physics supervision to an existing scene understanding model increases performance, produces more stable predictions, and allows training to an equivalent performance level with fewer annotated training examples. We then present a novel architecture for 3D scene parsing named Prim R-CNN, learning to predict bounding boxes as well as their 3D size, translation, and rotation. With physics supervision, Prim R-CNN outperforms existing scene understanding approaches on this problem. Finally, we show that applying physics supervision on unlabeled real images improves real domain transfer of models training on synthetic data. Yilun Du, Hector Basevi, Ales Leonardis, William T. Freeman, Josh Tenenbaum, Jiajun Wu 0001	NeurIPS	7
2018	3D-Aware Scene Manipulation via Inverse Graphics abstract We aim to obtain an interpretable, expressive, and disentangled scene representation that contains comprehensive structural and textural information for each object. Previous scene representations learned by neural networks are often uninterpretable, limited to a single object, or lacking 3D knowledge. In this work, we propose 3D scene de-rendering networks (3D-SDN) to address the above issues by integrating disentangled representations for semantics, geometry, and appearance into a deep generative model. Our scene encoder performs inverse graphics, translating a scene into a structured object-wise representation. Our decoder has two components: a differentiable shape renderer and a neural texture generator. The disentanglement of semantics, geometry, and appearance supports 3D-aware scene manipulation, e.g., rotating and moving objects freely while keeping the consistent shape and texture, and changing the object appearance without affecting its shape. Experiments demonstrate that our editing scheme based on 3D-SDN is superior to its 2D counterpart. Shunyu Yao 0006, Tzu-Ming Harry Hsu, Jun-Yan Zhu, Jiajun Wu 0001, Antonio Torralba 0001, William T. Freeman, Josh Tenenbaum	NeurIPS	4
2018	Neural-Symbolic VQA: Disentangling Reasoning from Vision and Language Understanding abstract We marry two powerful ideas: deep representation learning for visual recognition and language understanding, and symbolic program execution for reasoning. Our neural-symbolic visual question answering (NS-VQA) system first recovers a structural scene representation from the image and a program trace from the question. It then executes the program on the scene representation to obtain an answer. Incorporating symbolic structure as prior knowledge offers three unique advantages. First, executing programs on a symbolic space is more robust to long program traces; our model can solve complex reasoning tasks better, achieving an accuracy of 99.8% on the CLEVR dataset. Second, the model is more data- and memory-efficient: it performs well after learning on a small number of training data; it can also encode an image into a compact representation, requiring less storage than existing methods for offline question answering. Third, symbolic program execution offers full transparency to the reasoning process; we are thus able to interpret and diagnose each execution step. Kexin Yi, Jiajun Wu 0001, Chuang Gan 0001, Antonio Torralba 0001, Pushmeet Kohli, Josh Tenenbaum	NeurIPS	2
2018	Learning to Reconstruct Shapes from Unseen Classes abstract From a single image, humans are able to perceive the full 3D shape of an object by exploiting learned shape priors from everyday life. Contemporary single-image 3D reconstruction algorithms aim to solve this task in a similar fashion, but often end up with priors that are highly biased by training classes. Here we present an algorithm, Generalizable Reconstruction (GenRe), designed to capture more generic, class-agnostic shape priors. We achieve this with an inference network and training procedure that combine 2.5D representations of visible surfaces (depth and silhouette), spherical shape representations of both visible and non-visible surfaces, and 3D voxel-based representations, in a principled manner that exploits the causal structure of how 3D shapes give rise to 2D images. Experiments demonstrate that GenRe performs well on single-view shape reconstruction, and generalizes to diverse novel objects from categories not seen during training. Xiuming Zhang, Zhoutong Zhang, Chengkai Zhang, Josh Tenenbaum, William T. Freeman, Jiajun Wu 0001	NeurIPS	6
2018	Visual Object Networks: Image Generation with Disentangled 3D Representations abstract Recent progress in deep generative models has led to tremendous breakthroughs in image generation. While being able to synthesize photorealistic images, existing models lack an understanding of our underlying 3D world. Different from previous works built on 2D datasets and models, we present a new generative model, Visual Object Networks (VONs), synthesizing natural images of objects with a disentangled 3D representation. Inspired by classic graphics rendering pipelines, we unravel the image formation process into three conditionally independent factors---shape, viewpoint, and texture---and present an end-to-end adversarial learning framework that jointly models 3D shape and 2D texture. Our model first learns to synthesize 3D shapes that are indistinguishable from real shapes. It then renders the object's 2.5D sketches (i.e., silhouette and depth map) from its shape under a sampled viewpoint. Finally, it learns to add realistic textures to these 2.5D sketches to generate realistic images. The VON not only generates images that are more realistic than the state-of-the-art 2D image synthesis methods but also enables many 3D operations such as changing the viewpoint of a generated image, shape and texture editing, linear interpolation in texture and shape space, and transferring appearance across different objects and viewpoints. Jun-Yan Zhu, Zhoutong Zhang, Chengkai Zhang, Jiajun Wu 0001, Antonio Torralba 0001, Josh Tenenbaum, William T. Freeman	NeurIPS	4
2018	Unsupervised Learning of Latent Physical Properties Using Perception-Prediction Networks David Zheng, Vinson Luo, Jiajun Wu 0001, Josh Tenenbaum	UAI	3
2018	MoSculp: Interactive Visualization of Shape and Time abstract We present a system that visualizes complex human motion via 3D motion sculptures-a representation that conveys the 3D structure swept by a human body as it moves through space. Our system computes a motion sculpture from an input video, and then embeds it back into the scene in a 3D-aware fashion. The user may also explore the sculpture directly in 3D or physically print it. Our interactive interface allows users to customize the sculpture design, for example, by selecting materials and lighting conditions. To provide this end-to-end workflow, we introduce an algorithm that estimates a human's 3D geometry over time from a set of 2D images, and develop a 3D-aware image-based rendering approach that inserts the sculpture back into the original video. By automating the process, our system takes motion sculpture creation out of the realm of professional artists, and makes it applicable to a wide range of existing video material. By conveying 3D information to users, motion sculptures reveal space-time motion information that is difficult to perceive with the naked eye, and allow viewers to interpret how different parts of the object interact over time. We validate the effectiveness of motion sculptures with user studies, finding that our visualizations are more informative about motion than existing stroboscopic and space-time visualization methods. Xiuming Zhang, Tali Dekel, Tianfan Xue, Andrew Owens, Qiurui He 0001, Jiajun Wu 0001, Stefanie Mueller 0001, William T. Freeman	UIST	6
2018	Learning Sight from Sound: Ambient Sound Provides Supervision for Visual Learning Andrew Owens, Jiajun Wu 0001, Josh H. McDermott, William T. Freeman, Antonio Torralba 0001	Int. J. Comput. Vis.	2
2018	3D Interpreter Networks for Viewer-Centered Wireframe Modeling Jiajun Wu 0001, Tianfan Xue, Joseph J. Lim, Yuandong Tian, Josh Tenenbaum, Antonio Torralba 0001, William T. Freeman	Int. J. Comput. Vis.	1
2017	Neural Scene De-rendering abstract We study the problem of holistic scene understanding. We would like to obtain a compact, expressive, and interpretable representation of scenes that encodes information such as the number of objects and their categories, poses, positions, etc. Such a representation would allow us to reason about and even reconstruct or manipulate elements of the scene. Previous works have used encoder-decoder based neural architectures to learn image representations, however, representations obtained in this way are typically uninterpretable, or only explain a single object in the scene. In this work, we propose a new approach to learn an interpretable distributed representation of scenes. Our approach employs a deterministic rendering function as the decoder, mapping a naturally structured and disentangled scene description, which we named scene XML, to an image. By doing so, the encoder is forced to perform the inverse of the rendering operation (a.k.a. de-rendering) to transform an input image to the structured scene XML that the decoder used to produce the image. We use a object proposal based encoder that is trained by minimizing both the supervised prediction and the unsupervised reconstruction errors. Experiments demonstrate that our approach works well on scene de-rendering with two different graphics engines, and our learned representation can be easily adapted for a wide range of applications like image editing, inpainting, visual analogy-making, and image captioning. Jiajun Wu 0001, Josh Tenenbaum, Pushmeet Kohli	CVPR	1
2017	Synthesizing 3D Shapes via Modeling Multi-view Depth Maps and Silhouettes with Deep Generative Networks abstract We study the problem of learning generative models of 3D shapes. Voxels or 3D parts have been widely used as the underlying representations to build complex 3D shapes, however, voxel-based representations suffer from high memory requirements, and parts-based models require a large collection of cached or richly parametrized parts. We take an alternative approach: learning a generative model over multi-view depth maps or their corresponding silhouettes, and using a deterministic rendering function to produce 3D shapes from these images. A multi-view representation of shapes enables generation of 3D models with fine details, as 2D depth maps and silhouettes can be modeled at a much higher resolution than 3D voxels. Moreover, our approach naturally brings the ability to recover the underlying 3D representation from depth maps of one or a few viewpoints. Experiments show that our framework can generate 3D shapes with variations and details. We also demonstrate that our model has out-of-sample generalization power for real-world tasks with occluded objects. Amir Arsalan Soltani, Jiajun Wu 0001, Tejas D. Kulkarni, Josh Tenenbaum	CVPR	3
2017	Raster-to-Vector: Revisiting Floorplan Transformation abstract This paper addresses the problem of converting a rasterized floorplan image into a vector-graphics representation. Unlike existing approaches that rely on a sequence of lowlevel image processing heuristics, we adopt a learning-based approach. A neural architecture first transforms a rasterized image to a set of junctions that represent low-level geometric and semantic information (e.g., wall corners or door end-points). Integer programming is then formulated to aggregate junctions into a set of simple primitives (e.g., wall lines, door lines, or icon boxes) to produce a vectorized floorplan, while ensuring a topologically and geometrically consistent result. Our algorithm significantly outperforms existing methods and achieves around 90% precision and recall, getting to the range of production-ready performance. The vector representation allows 3D model popup for better indoor scene visualization, direct model manipulation for architectural remodeling, and further computational applications such as data analysis. Our system is efficient: we have converted hundred thousand production-level floorplan images into the vector representation and generated 3D popup models. Chen Liu 0012, Jiajun Wu 0001, Pushmeet Kohli, Yasutaka Furukawa	ICCV	2
2017	Generative Modeling of Audible Shapes for Object Perception abstract Humans infer rich knowledge of objects from both auditory and visual cues. Building a machine of such competency, however, is very challenging, due to the great difficulty in capturing large-scale, clean data of objects with both their appearance and the sound they make. In this paper, we present a novel, open-source pipeline that generates audiovisual data, purely from 3D object shapes and their physical properties. Through comparison with audio recordings and human behavioral studies, we validate the accuracy of the sounds it generates. Using this generative model, we are able to construct a synthetic audio-visual dataset, namely Sound-20K, for object perception tasks. We demonstrate that auditory and visual information play complementary roles in object perception, and further, that the representation learned on synthetic audio-visual data can transfer to real-world scenarios. Zhoutong Zhang, Jiajun Wu 0001, Qiujia Li, Zhengjia Huang, James Traer, Josh H. McDermott, Josh Tenenbaum, William T. Freeman	ICCV	2
2017	Learning to See Physics via Visual De-animation abstract We introduce a paradigm for understanding physical scenes without human annotations. At the core of our system is a physical world representation that is first recovered by a perception module and then utilized by physics and graphics engines. During training, the perception module and the generative models learn by visual de-animation --- interpreting and reconstructing the visual information stream. During testing, the system first recovers the physical world state, and then uses the generative models for reasoning and future prediction. Even more so than forward simulation, inverting a physics or graphics engine is a computationally hard problem; we overcome this challenge by using a convolutional inversion network. Our system quickly recognizes the physical world state from appearance and motion cues, and has the flexibility to incorporate both differentiable and non-differentiable physics and graphics engines. We evaluate our system on both synthetic and real datasets involving multiple physical scenes, and demonstrate that our system performs well on both physical state estimation and reasoning problems. We further show that the knowledge learned on the synthetic dataset generalizes to constrained real images. Jiajun Wu 0001, Erika Lu, Pushmeet Kohli, William T. Freeman, Josh Tenenbaum	NIPS	1
2017	MarrNet: 3D Shape Reconstruction via 2.5D Sketches abstract 3D object reconstruction from a single image is a highly under-determined problem, requiring strong prior knowledge of plausible 3D shapes. This introduces challenge for learning-based approaches, as 3D object annotations in real images are scarce. Previous work chose to train on synthetic data with ground truth 3D information, but suffered from the domain adaptation issue when tested on real data. In this work, we propose an end-to-end trainable framework, sequentially estimating 2.5D sketches and 3D object shapes. Our disentangled, two-step formulation has three advantages. First, compared to full 3D shape, 2.5D sketches are much easier to be recovered from a 2D image, and to transfer from synthetic to real data. Second, for 3D reconstruction from the 2.5D sketches, we can easily transfer the learned model on synthetic data to real images, as rendered 2.5D sketches are invariant to object appearance variations in real images, including lighting, texture, etc. This further relieves the domain adaptation problem. Third, we derive differentiable projective functions from 3D shape to 2.5D sketches, making the framework end-to-end trainable on real images, requiring no real-image annotations. Our framework achieves state-of-the-art performance on 3D shape reconstruction. Jiajun Wu 0001, Wang Yifan 0001, Tianfan Xue, Xingyuan Sun, William T. Freeman, Josh Tenenbaum	NIPS	1
2017	Self-Supervised Intrinsic Image Decomposition abstract Intrinsic decomposition from a single image is a highly challenging task, due to its inherent ambiguity and the scarcity of training data. In contrast to traditional fully supervised learning approaches, in this paper we propose learning intrinsic image decomposition by explaining the input image. Our model, the Rendered Intrinsics Network (RIN), joins together an image decomposition pipeline, which predicts reflectance, shape, and lighting conditions given a single image, with a recombination function, a learned shading model used to recompose the original input based off of intrinsic image predictions. Our network can then use unsupervised reconstruction error as an additional signal to improve its intermediate representations. This allows large-scale unlabeled data to be useful during training, and also enables transferring learned knowledge to images of unseen object categories, lighting conditions, and shapes. Extensive experiments demonstrate that our method performs well on both intrinsic image decomposition and knowledge transfer. Michael Janner, Jiajun Wu 0001, Tejas D. Kulkarni, Ilker Yildirim, Josh Tenenbaum	NIPS	2
2017	Shape and Material from Sound abstract Hearing an object falling onto the ground, humans can recover rich information including its rough shape, material, and falling height. In this paper, we build machines to approximate such competency. We first mimic human knowledge of the physical world by building an efficient, physics-based simulation engine. Then, we present an analysis-by-synthesis approach to infer properties of the falling object. We further accelerate the process by learning a mapping from a sound wave to object properties, and using the predicted values to initialize the inference. This mapping can be viewed as an approximation of human commonsense learned from past experience. Our model performs well on both synthetic audio clips and real recordings without requiring any annotated data. We conduct behavior studies to compare human responses with ours on estimating object shape, material, and falling height from sound. Our model achieves near-human performance. Zhoutong Zhang, Qiujia Li, Zhengjia Huang, Jiajun Wu 0001, Josh Tenenbaum, William T. Freeman	NIPS	4
2016	Physics 101: Learning Physical Object Properties from Unlabeled Videos Jiajun Wu 0001, Joseph J. Lim, Josh Tenenbaum, William T. Freeman	BMVC	1
2016	A Comparative Evaluation of Approximate Probabilistic Simulation and Deep Neural Networks as Accounts of Human Physical Scene Understanding Renqiao Zhang, Jiajun Wu 0001, Chengkai Zhang, William T. Freeman, Josh Tenenbaum	CogSci	2
2016	Single Image 3D Interpreter Network Jiajun Wu 0001, Tianfan Xue, Joseph J. Lim, Yuandong Tian, Josh Tenenbaum, Antonio Torralba 0001, William T. Freeman	ECCV (6)	1
2016	Ambient Sound Provides Supervision for Visual Learning Andrew Owens, Jiajun Wu 0001, Josh H. McDermott, William T. Freeman, Antonio Torralba 0001	ECCV (1)	2
2016	Learning a Probabilistic Latent Space of Object Shapes via 3D Generative-Adversarial Modeling abstract We study the problem of 3D object generation. We propose a novel framework, namely 3D Generative Adversarial Network (3D-GAN), which generates 3D objects from a probabilistic space by leveraging recent advances in volumetric convolutional networks and generative adversarial nets. The benefits of our model are three-fold: first, the use of an adversarial criterion, instead of traditional heuristic criteria, enables the generator to capture object structure implicitly and to synthesize high-quality 3D objects; second, the generator establishes a mapping from a low-dimensional probabilistic space to the space of 3D objects, so that we can sample objects without a reference image or CAD models, and explore the 3D object manifold; third, the adversarial discriminator provides a powerful 3D shape descriptor which, learned without supervision, has wide applications in 3D object recognition. Experiments demonstrate that our method generates high-quality 3D objects, and our unsupervisedly learned features achieve impressive performance on 3D object recognition, comparable with those of supervised learning methods. Jiajun Wu 0001, Chengkai Zhang, Tianfan Xue, William T. Freeman, Josh Tenenbaum	NIPS	1
2016	Visual Dynamics: Probabilistic Future Frame Synthesis via Cross Convolutional Networks abstract We study the problem of synthesizing a number of likely future frames from a single input image. In contrast to traditional methods, which have tackled this problem in a deterministic or non-parametric way, we propose a novel approach which models future frames in a probabilistic manner. Our proposed method is therefore able to synthesize multiple possible next frames using the same model. Solving this challenging problem involves low- and high-level image and motion understanding for successful image synthesis. Here, we propose a novel network structure, namely a Cross Convolutional Network, that encodes images as feature maps and motion information as convolutional kernels to aid in synthesizing future frames. In experiments, our model performs well on both synthetic data, such as 2D shapes and animated game sprites, as well as on real-wold video data. We show that our model can also be applied to tasks such as visual analogy-making, and present analysis of the learned network representations. Tianfan Xue, Jiajun Wu 0001, Katherine L. Bouman, William T. Freeman	NIPS	2
2015	Deep multiple instance learning for image classification and auto-annotation abstract The recent development in learning deep representations has demonstrated its wide applications in traditional vision tasks like classification and detection. However, there has been little investigation on how we could build up a deep learning framework in a weakly supervised setting. In this paper, we attempt to model deep learning in a weakly supervised learning (multiple instance learning) framework. In our setting, each image follows a dual multi-instance assumption, where its object proposals and possible text annotations can be regarded as two instance sets. We thus design effective systems to exploit the MIL property with deep learning strategies from the two ends; we also try to jointly learn the relationship between object and annotation proposals. We conduct extensive experiments and prove that our weakly supervised deep learning framework not only achieves convincing performance in vision tasks including classification and image annotation, but also extracts reasonable region-keyword pairs with little supervision, on both widely used benchmarks like PASCAL VOC and MIT Indoor Scene 67, and also a dataset for image-and patch-level annotations. Jiajun Wu 0001, Yinan Yu, Chang Huang	CVPR	1
2015	Galileo: Perceiving Physical Object Properties by Integrating a Physics Engine with Deep Learning abstract Humans demonstrate remarkable abilities to predict physical events in dynamic scenes, and to infer the physical properties of objects from static images. We propose a generative model for solving these problems of physical scene understanding from real-world videos and images. At the core of our generative model is a 3D physics engine, operating on an object-based representation of physical properties, including mass, position, 3D shape, and friction. We can infer these latent properties using relatively brief runs of MCMC, which drive simulations in the physics engine to fit key features of visual observations. We further explore directly mapping visual inputs to physical properties, inverting a part of the generative process using deep learning. We name our model Galileo, and evaluate it on a video dataset with simple yet physically rich scenarios. Results show that Galileo is able to infer the physical properties of objects and predict the outcome of a variety of physical events, with an accuracy comparable to human subjects. Our study points towards an account of human vision with generative physical knowledge at its core, and various recognition models as helpers leading to efficient inference. Jiajun Wu 0001, Ilker Yildirim, Joseph J. Lim, William T. Freeman, Josh Tenenbaum	NIPS	1
2015	Unsupervised Object Class Discovery via Saliency-Guided Multiple Class Learning abstract In this paper, we tackle the problem of common object (multiple classes) discovery from a set of input images, where we assume the presence of one object class in each image. This problem is, loosely speaking, unsupervised since we do not know a priori about the object type, location, and scale in each image. We observe that the general task of object class discovery in a fully unsupervised manner is intrinsically ambiguous; here we adopt saliency detection to propose candidate image windows/patches to turn an unsupervised learning problem into a weakly-supervised learning problem. In the paper, we propose an algorithm for simultaneously localizing objects and discovering object classes via bottom-up (saliency-guided) multiple class learning (bMCL). Our contributions are three-fold: (1) we adopt saliency detection to convert unsupervised learning into multiple instance learning, formulated as bottom-up multiple class learning (bMCL); (2) we propose an integrated framework that simultaneously performs object localization, object class discovery, and object detector training; (3) we demonstrate that our framework yields significant improvements over existing methods for multi-class object discovery and possess evident advantages over competing methods in computer vision. In addition, although saliency detection has recently attracted much attention, its practical usage for high-level vision tasks has yet to be justified. Our method validates the usefulness of saliency detection to output "noisy input" for a top-down method to extract common patterns. Jun-Yan Zhu, Jiajun Wu 0001, Yan Xu 0001, Eric I-Chao Chang, Zhuowen Tu	IEEE Trans. Pattern Anal. Mach. Intell.	2
2014	Reverse Image Segmentation: A High-Level Solution to a Low-Level Task Jiajun Wu 0001, Junyan Zhu, Zhuowen Tu	BMVC	1
2014	Harvesting Motion Patterns in Still Images from the Internet Jiajun Wu 0001, Zhulin Li, Zhuowen Tu	CogSci	1
2014	MILCut: A Sweeping Line Multiple Instance Learning Paradigm for Interactive Image Segmentation abstract Interactive segmentation, in which a user provides a bounding box to an object of interest for image segmentation, has been applied to a variety of applications in image editing, crowdsourcing, computer vision, and medical imaging. The challenge of this semi-automatic image segmentation task lies in dealing with the uncertainty of the foreground object within a bounding box. Here, we formulate the interactive segmentation problem as a multiple instance learning (MIL) task by generating positive bags from pixels of sweeping lines within a bounding box. We name this approach MILCut. We provide a justification to our formulation and develop an algorithm with significant performance and efficiency gain over existing state-of-the-art systems. Extensive experiments demonstrate the evident advantage of our approach. Jiajun Wu 0001, Yibiao Zhao, Jun-Yan Zhu, Siwei Luo, Zhuowen Tu	CVPR	1
2013	Harvesting Mid-level Visual Concepts from Large-Scale Internet Images abstract Obtaining effective mid-level representations has become an increasingly important task in computer vision. In this paper, we propose a fully automatic algorithm which harvests visual concepts from a large number of Internet images (more than a quarter of a million) using text-based queries. Existing approaches to visual concept learning from Internet images either rely on strong supervision with detailed manual annotations or learn image-level classifiers only. Here, we take the advantage of having massive well organized Google and Bing image data, visual concepts (around 14, 000) are automatically exploited from images using word-based queries. Using the learned visual concepts, we show state-of-the-art performances on a variety of benchmark datasets, which demonstrate the effectiveness of the learned mid-level representations: being able to generalize well to general natural images. Our method shows significant improvement over the competing systems in image classification, including those with strong supervision. Quannan Li, Jiajun Wu 0001, Zhuowen Tu	CVPR	2
2012	Unsupervised object class discovery via saliency-guided multiple class learning abstract Discovering object classes from images in a fully unsupervised way is an intrinsically ambiguous task; saliency detection approaches however ease the burden on unsupervised learning. We develop an algorithm for simultaneously localizing objects and discovering object classes via bottom-up (saliency-guided) multiple class learning (bMCL), and make the following contributions: (1) saliency detection is adopted to convert unsupervised learning into multiple instance learning, formulated as bottom-up multiple class learning (bMCL); (2) we utilize the Discriminative EM (DiscEM) to solve our bMCL problem and show DiscEM's connection to the MIL-Boost method[34]; (3) localizing objects, discovering object classes, and training object detectors are performed simultaneously in an integrated framework; (4) significant improvements over the existing methods for multi-class object discovery are observed. In addition, we show single class localization as a special case in our bMCL framework and we also demonstrate the advantage of bMCL over purely data-driven saliency methods. Jun-Yan Zhu, Jiajun Wu 0001, Eric I-Chao Chang, Zhuowen Tu	CVPR	2
2012	A classification approach to coreference in discharge summaries: 2011 i2b2 challenge abstract OBJECTIVE: To create a highly accurate coreference system in discharge summaries for the 2011 i2b2 challenge. The coreference categories include Person, Problem, Treatment, and Test. DESIGN: An integrated coreference resolution system was developed by exploiting Person attributes, contextual semantic clues, and world knowledge. It includes three subsystems: Person coreference system based on three Person attributes, Problem/Treatment/Test system based on numerous contextual semantic extractors and world knowledge, and Pronoun system based on a multi-class support vector machine classifier. The three Person attributes are patient, relative and hospital personnel. Contextual semantic extractors include anatomy, position, medication, indicator, temporal, spatial, section, modifier, equipment, operation, and assertion. The world knowledge is extracted from external resources such as Wikipedia. MEASUREMENTS: Micro-averaged precision, recall and F-measure in MUC, BCubed and CEAF were used to evaluate results. RESULTS: The system achieved an overall micro-averaged precision, recall and F-measure of 0.906, 0.925, and 0.915, respectively, on test data (from four hospitals) released by the challenge organizers. It achieved a precision, recall and F-measure of 0.905, 0.920 and 0.913, respectively, on test data without Pittsburgh data. We ranked the first out of 20 competing teams. Among the four sub-tasks on Person, Problem, Treatment, and Test, the highest F-measure was seen for Person coreference. CONCLUSIONS: This system achieved encouraging results. The Person system can determine whether personal pronouns and proper names are coreferent or not. The Problem/Treatment/Test system benefits from both world knowledge in evaluating the similarity of two mentions and contextual semantic extractors in identifying semantic clues. The Pronoun system can automatically detect whether a Pronoun mention is coreferent to that of the other four types. This study demonstrates that it is feasible to accomplish the coreference task in discharge summaries. Yan Xu 0001, Jiahua Liu, Jiajun Wu 0001, Yue Wang 0035, Zhuowen Tu, Jian-Tao Sun, Jun'ichi Tsujii, Eric I-Chao Chang	J. Am. Medical Informatics Assoc.	3