Efficiently reconstructing 3D scenes from monocular video remains a core challenge in computer vision, vital for applications in virtual reality, robotics, and scene understanding. Recently, frame-by-frame progressive reconstruction without camera poses is commonly adopted, incurring high computational overhead and compounding errors when scaling to longer videos. To overcome these issues, we introduce VideoLifter, a novel video-to-3D pipeline that leverages a local-to-global strategy on a fragment basis, achieving both extreme efficiency and SOTA quality. Locally, VideoLifter leverages learnable 3D priors to register fragments, extracting essential information for subsequent 3D Gaussian initialization with enforced inter-fragment consistency and optimized efficiency. Globally, it employs a tree-based hierarchical merging method with key frame guidance for inter-fragment alignment, pairwise merging with Gaussian point pruning, and subsequent joint optimization to ensure global consistency while efficiently mitigating cumulative errors. This approach significantly accelerates the reconstruction process, reducing training time by over 82 % while achieving better visual quality than SOTA methods.

Score distillation has emerged as one of the most prevalent approaches for text-to-3D asset synthesis. Essentially, score distillation updates 3D parameters by lifting and back-propagating scores averaged over different views. In this paper, we reveal that the gradient estimation in score distillation is inherent to high variance. Through the lens of variance reduction, the effectiveness of SDS and VSD can be interpreted as applications of various control variates to the Monte Carlo estimator of the distilled score. Motivated by this rethinking and based on Stein’s identity, we propose a more general solution to reduce variance for score distillation, termed \emph{Stein Score Distillation (SSD)}. SSD incorporates control variates constructed by Stein identity, allowing for arbitrary baseline functions. This enables us to include flexible guidance priors and network architectures to explicitly optimize for variance reduction. In our experiments, the overall pipeline, dubbed \emph{SteinDreamer}, is implemented by instantiating the control variate with a monocular depth estimator. The results show that SSD can effectively reduce the distillation variance and consistently improve visual quality for both object- and scene-level generation.

Recent advancements in 2D and multimodal models have achieved remarkable success by leveraging large-scale training on extensive datasets. However, extending these achievements to enable free-form interactions and high-level semantic operations with complex 3D/4D scenes remains challenging. This difficulty stems from the limited availability of large-scale, annotated 3D/4D or multi-view datasets, which are crucial for generalizable vision and language tasks such as open-vocabulary and prompt-based segmentation, language-guided editing, and visual question answering (VQA). In this paper, we introduce Feature4X, a universal framework designed to extend any functionality from 2D vision foundation model into the 4D realm, using only monocular video input, which is widely available from user-generated content. The "X" in Feature4X represents its versatility, enabling any task through adaptable, model-conditioned 4D feature field distillation. At the core of our framework is a dynamic optimization strategy that unifies multiple model capabilities into a single representation. Additionally, to the best of our knowledge, Feature4X is the first method to distill and lift the features of video foundation models (e.g. SAM2, InternVideo2) into an explicit 4D feature field using Gaussian Splatting. Our experiments showcase novel view segment anything, geometric and appearance scene editing, and free-form VQA across all time steps, empowered by LLMs in feedback loops. These advancements broaden the scope of agentic AI applications by providing a foundation for scalable, contextually and spatiotemporally aware systems capable of immersive dynamic 4D scene interaction.

3D Gaussian Splatting (3DGS) has emerged as a powerful technique for real-time, high-resolution novel view synthesis. By representing scenes as a mixture of Gaussian primitives, 3DGS leverages GPU rasterization pipelines for efficient rendering and reconstruction. To optimize scene coverage and capture fine details, 3DGS employs a densification algorithm to generate additional points. However, this process often leads to redundant point clouds, resulting in excessive memory usage, slower performance, and substantial storage demands–posing significant challenges for deployment on resource-constrained devices. To address this limitation, we propose a theoretical framework that demystifies and improves density control in 3DGS. Our analysis reveals that splitting is crucial for escaping saddle points. Through an optimization-theoretic approach, we establish the necessary conditions for densification, determine the minimal number of offspring Gaussians, identify the optimal parameter update direction, and provide an analytical solution for normalizing off-spring opacity. Building on these insights, we introduce SteepGS, incorporating steepest density control, a principled strategy that minimizes loss while maintaining a compact point cloud. SteepGS achieves a ~ 50% reduction in Gaussian points without compromising rendering quality, significantly enhancing both efficiency and scalability.

Pose estimation from unordered images is fundamental for 3D reconstruction, robotics, and scientific imaging. Recent geometric foundation models, such as DUSt3R, enable end-to-end dense 3D reconstruction but remain underexplored in scientific imaging fields like cryo-electron microscopy (cryo-EM) for near-atomic protein reconstruction. In cryo-EM, pose estimation and 3D reconstruction from unordered particle images still depend on time-consuming iterative optimization, primarily due to challenges such as low signal-to-noise ratios (SNR) and distortions from the contrast transfer function (CTF). We introduce CryoFastAR, the first geometric foundation model that can directly predict poses from Cryo-EM noisy images for Fast ab initio Reconstruction. By integrating multi-view features and training on large-scale simulated cryo-EM data with realistic noise and CTF modulations, CryoFastAR enhances pose estimation accuracy and generalization. To enhance training stability, we propose a progressive training strategy that first allows the model to extract essential features under simpler conditions before gradually increasing difficulty to improve robustness. Experiments show that CryoFastAR achieves comparable quality while significantly accelerating inference over traditional iterative approaches on both synthetic and real datasets.

Perception is a crucial component of autonomous driving systems. However, single-agent setups often face limitations due to sensor constraints, especially under challenging conditions like severe occlusion, adverse weather, and long-range object detection. Multi-agent collaborative perception (CP) offers a promising solution that enables communication and information sharing between connected vehicles. Yet, the heterogeneity among agents—in terms of sensors, models, and tasks—significantly hinders effective and efficient cross-agent collaboration. To address these challenges, we propose STAMP, a scalable task- and model-agnostic collaborative perception framework tailored for heterogeneous agents. STAMP utilizes lightweight adapter-reverter pairs to transform Bird's Eye View (BEV) features between agent-specific domains and a shared protocol domain, facilitating efficient feature sharing and fusion while minimizing computational overhead. Moreover, our approach enhances scalability, preserves model security, and accommodates a diverse range of agents. Extensive experiments on both simulated (OPV2V) and real-world (V2V4Real) datasets demonstrate that STAMP achieves comparable or superior accuracy to state-of-the-art models with significantly reduced computational costs. As the first-of-its-kind task- and model-agnostic collaborative perception framework, STAMP aims to advance research in scalable and secure mobility systems, bringing us closer to Level 5 autonomy. Our project page is at https://xiangbogaobarry.github.io/STAMP and the code is available at https://github.com/taco-group/STAMP.

With the rapid development of large generative models for 3D, especially the evolution from NeRF representations to more efficient Gaussian Splatting, the synthesis of 3D assets has become increasingly fast and efficient, enabling the large-scale publication and sharing of generated 3D objects. However, while existing methods can add watermarks or steganographic information to individual 3D assets, they often require time-consuming per-scene training and optimization, leading to watermarking overheads that can far exceed the time required for asset generation itself, making deployment impractical for generating large collections of 3D objects. To address this, we propose InstantSplamp a framework that seamlessly integrates the 3D steganography pipeline into large 3D generative models without introducing explicit additional time costs. Guided by visual foundation models,InstantSplamp subtly injects hidden information like copyright tags during asset generation, enabling effective embedding and recovery of watermarks within generated 3D assets while preserving original visual quality. Experiments across various potential deployment scenarios demonstrate that \model~strikes an optimal balance between rendering quality and hiding fidelity, as well as between hiding performance and speed. Compared to existing per-scene optimization techniques for 3D assets, InstantSplamp reduces their watermarking training overheads that are multiples of generation time to nearly zero, paving the way for real-world deployment at scale. Project page: https://gaussian-stego.github.io/.

The blooming of virtual reality and augmented reality (VR/AR) technologies has driven an increasing demand for the creation of high-quality, immersive, and dynamic environments. However, existing generative techniques either focus solely on dynamic objects or perform outpainting from a single perspective image, failing to meet the requirements of VR/AR applications that need free-viewpoint, 360$^{\circ}$ virtual views where users can move in all directions. In this work, we tackle the challenging task of elevating a single panorama to an immersive 4D experience. For the first time, we demonstrate the capability to generate omnidirectional dynamic scenes with 360$^{\circ}$ views at 4K (4096 $\times$ 2048) resolution, thereby providing an immersive user experience. Our method introduces a pipeline that facilitates natural scene animations and optimizes a set of 3D Gaussians using efficient splatting techniques for real-time exploration. To overcome the lack of scene-scale annotated 4D data and models, especially in panoramic formats, we propose a novel Panoramic Denoiser that adapts generic 2D diffusion priors to animate consistently in 360$^{\circ}$ images, transforming them into panoramic videos with dynamic scenes at targeted regions. Subsequently, we propose Dynamic Panoramic Lifting to elevate the panoramic video into a 4D immersive environment while preserving spatial and temporal consistency. By transferring prior knowledge from 2D models in the perspective domain to the panoramic domain and the 4D lifting with spatial appearance and geometry regularization, we achieve high-quality Panorama-to-4D generation at a resolution of 4K for the first time. Project page: https://4k4dgen.github.io/.

The synthesis of realistic Martian landscape videos, essential for mission rehearsal and robotic simulation, presents unique challenges. These primarily stem from the scarcity of high-quality Martian data and the significant domain gap relative to terrestrial imagery.To address these challenges, we introduce a holistic solution comprising two main components: 1) a data curation framework, Multimodal Mars Synthesis (M3arsSynth), which processes stereo navigation images to render high-fidelity 3D video sequences. 2) a video-based Martian terrain generator (MarsGen), that utilizes multimodal conditioning data to accurately synthesize novel, 3D-consistent frames. Our data are sourced from NASA’s Planetary Data System (PDS), covering diverse Martian terrains and dates, enabling the production of physics-accurate 3D surface models at metric-scale resolution. During inference, MarsGen is conditioned on an initial image frame and can be guided by specified camera trajectories or textual prompts to generate new environments.Experimental results demonstrate that our solution surpasses video synthesis approaches trained on terrestrial data, achieving superior visual quality and 3D structural consistency.

Understanding the dynamic physical world, characterized by its evolving 3D structure, real-world motion, and semantic content with textual descriptions, is crucial for human-agent interaction and enables embodied agents to perceive and act within real environments with human‑like capabilities. However, existing datasets are often derived from limited simulators or utilize traditional Structure-from-Motion for up-to-scale annotation and offer limited descriptive captioning, which restricts the capacity of foundation models to accurately interpret real-world dynamics from monocular videos, commonly sourced from the internet. To bridge these gaps, we introduce **DynamicVerse**, a physical‑scale, multimodal 4D world modeling framework for dynamic real-world video. We employ large vision, geometric, and multimodal models to interpret metric-scale static geometry, real-world dynamic motion, instance-level masks, and holistic descriptive captions. By integrating window-based Bundle Adjustment with global optimization, our method converts long real-world video sequences into a comprehensive 4D multimodal format. DynamicVerse delivers a large-scale dataset consists of 100K+ videos with 800K+ annotated masks and 10M+ frames from internet videos. Experimental evaluations on three benchmark tasks, namely video depth estimation, camera pose estimation, and camera intrinsics estimation, demonstrate that our 4D modeling achieves superior performance in capturing physical-scale measurements with greater global accuracy than existing methods.

Learning efficient and interpretable policies has been a challenging task in reinforcement learning (RL), particularly in the visual RL setting with complex scenes. While neural networks have achieved competitive performance, the resulting policies are often over-parameterized black boxes that are difficult to interpret and deploy efficiently. More recent SRL frameworks have shown that high-level domain-specific programming logic can be designed to handle both policy learning and symbolic planning. However, these approaches rely on coded primitives with little feature learning, and when applied to high-dimensional visual scenes, they can suffer from scalability issues and perform poorly when images have complex object interactions. To address these challenges, we propose Differentiable Symbolic Expression Search (DiffSES), a novel symbolic learning approach that discovers discrete symbolic policies using partially differentiable optimization. By using object-level abstractions instead of raw pixel-level inputs, DiffSES is able to leverage the simplicity and scalability advantages of symbolic expressions, while also incorporating the strengths of neural networks for feature learning and optimization. Our experiments demonstrate that DiffSES is able to generate symbolic policies that are simpler and more and scalable than state-of-the-art SRL methods, with a reduced amount of symbolic prior knowledge.

The accurate estimation of six degrees-of-freedom (6DoF) object poses is essential for many applications in robotics and augmented reality. However, existing methods for 6DoF pose estimation often depend on CAD templates or dense support views, restricting their usefulness in real-world situations. In this study, we present a new cascade framework named Cas6D for few-shot 6DoF pose estimation that is generalizable and uses only RGB images. To address the false positives of target object detection in the extreme few-shot setting, our framework utilizes a self-supervised pre-trained ViT to learn robust feature representations. Then, we initialize the nearest top-K pose candidates based on similarity score and refine the initial poses using feature pyramids to formulate and update the cascade warped feature volume, which encodes context at increasingly finer scales. By discretizing the pose search range using multiple pose bins and progressively narrowing the pose search range in each stage using predictions from the previous stage, Cas6D can overcome the large gap between pose candidates and ground truth poses, which is a common failure mode in sparse-view scenarios. Experimental results on the LINEMOD and GenMOP datasets demonstrate that Cas6D outperforms state-of-the-art methods by 9.2% and 3.8& accuracy (Proj-5) under the 32-shot setting compared to OnePose++ and Gen6D. Our framework also performs best under the full-shot setting with all support views. Code is avilable at https://github.com/github.com/paulpanwang/Cas6D.

In recent years, there has been an explosion of 2D vision models for numerous tasks such as semantic segmentation, style transfer or scene editing, enabled by large-scale 2D image datasets. At the same time, there has been renewed interest in 3D scene representations such as neural radiance fields from multi-view images. However, the availability of 3D or multiview data is still substantially limited compared to 2D image datasets, making extending 2D vision models to 3D data highly desirable but also very challenging. Indeed, extending a single 2D vision operator like scene editing to 3D typically requires a highly creative method specialized to that task and often requires per-scene optimization. In this paper, we ask the question of whether any 2D vision model can be lifted to make 3D consistent predictions. We answer this question in the affirmative; our new Lift3D method trains to predict unseen views onfeature spaces generated by afew visual models (i.e. DINO and CLIP), but then generalizes to novel vision operators and tasks, such as style transfer, super-resolution, open vocabulary segmentation and image colorization; for some of these tasks, there is no comparable previous 3D method. In many cases, we even outperform state-of-the-art methods specialized for the task in question. Moreover, Lift3D is a zero-shot method, in the sense that it requires no task-specific training, nor scene-specific optimization.

Despite the remarkable performance of score distillation in text-to-3D generation, such techniques notoriously suf-fer from view inconsistency issues, also known as “Janus” artifact, where the generated objects fake each view with multiple front faces. Although empirically effective methods have approached this problem via score debiasing or prompt engineering, a more rigorous perspective to explain and tackle this problem remains elusive. In this paper, we reveal that the existing score distillation-based text-to-3D generation frameworks degenerate to maximal likelihood seeking on each view independently and thus suffer from the mode collapse problem, manifesting as the Janus artifact in practice. To tame mode collapse, we improve score distillation by re-establishing the entropy term in the corresponding variational objective, which is applied to the distribution of rendered images. Maximizing the entropy encourages diversity among different views in generated 3D assets, thereby mitigating the Janus problem. Based on this new objective, we derive a new update rule for 3D score distillation, dubbed Entropic Score Distillation (ESD). We theoretically reveal that ESD can be simplified and implemented by just adopting the classifier-free guidance trick upon variational score distillation. Although embarrassingly straightforward, our extensive experiments demonstrate that ESD can be an effective treatment for Janus artifacts in score distillation.

3D scene representations have gained immense popularity in recent years. Methods that use Neural Radiance fields are versatile for traditional tasks such as novel view synthesis. In recent times, some work has emerged that aims to extend the functionality of NeRF beyond view synthesis, for semantically aware tasks such as editing and segmentation using 3D feature field distillation from 2D foundation models. However, these methods have two major limitations: (a) they are limited by the rendering speed of NeRF pipelines, and (b) implicitly represented feature fields suffer from continuity artifacts reducing feature quality. Recently, 3D Gaussian Splatting has shown state-of-the-art performance on real-time radiance field rendering. In this work, we go one step further: in addition to radiance field rendering, we enable 3D Gaussian splatting on arbitrary-dimension semantic features via 2D foundation model distillation. This translation is not straightforward: naively incorporating feature fields in the 3DGS framework encounters significant challenges, notably the disparities in spatial resolution and channel consistency between RGB images and feature maps. We propose architectural and training changes to efficiently avert this problem. Our proposed method is general, and our experiments showcase novel view semantic segmentation, language-guided editing and segment anything through learning feature fields from state-of-the-art 2D foundation models such as SAM and CLIP-LSeg. Across experiments, our distillation method is able to provide comparable or better results, while being significantly faster to both train and render. Additionally, to the best of our knowledge, we are the first method to enable point and bounding-box prompting for radiance field manipulation, by leveraging the SAM model. Project website at: https://feature-3dgs.github.io/.

Simultaneous localization and mapping is essential for position tracking and scene understanding. 3D Gaussian-based map representations enable photorealistic reconstruction and real-time rendering of scenes using multiple posed cameras. We show for the first time that using 3D Gaussians for map representation with unposed camera images and inertial measurements can enable accurate SLAM. Our method, MM3DGS, addresses the limitations of prior neural radiance field-based representations by enabling faster rendering, scale awareness, and improved trajectory tracking. Our framework enables keyframe-based mapping and tracking utilizing loss functions that incorporate relative pose transformations from pre-integrated inertial measurements, depth estimates, and measures of photometric rendering quality. We also release a multi-modal dataset, UT-MM, collected from a mobile robot equipped with a camera and an inertial measurement unit. Experimental evaluation on several scenes from the dataset shows that MM3DGS achieves nearly 3x improvement in tracking and 5% improvement in photometric rendering quality compared to the current 3DGS SLAM state-of-the-art, while allowing real-time rendering of a high-resolution dense 3D map.

Recent advances in real-time neural rendering using point-based techniques have enabled broader adoption of 3D representations. However, foundational approaches like 3D Gaussian Splatting impose substantial storage overhead, as Structure-from-Motion (SfM) points can grow to millions, often requiring gigabyte-level disk space for a single unbounded scene. This growth presents scalability challenges and hinders splatting efficiency. To address this, we introduce LightGaussian, a method for transforming 3D Gaussians into a more compact format. Inspired by Network Pruning, LightGaussian identifies Gaussians with minimal global significance on scene reconstruction, and applies a pruning and recovery process to reduce redundancy while preserving visual quality. Knowledge distillation and pseudo-view augmentation then transfer spherical harmonic coefficients to a lower degree, yielding compact representations. Gaussian Vector Quantization, based on each Gaussian’s global significance, further lowers bitwidth with minimal accuracy loss. LightGaussian achieves an average 15 times compression rate while boosting FPS from 144 to 237 within the 3D-GS framework, enabling efficient complex scene representation on the Mip-NeRF 360 and Tank & Temple datasets. The proposed Gaussian pruning approach is also adaptable to other 3D representations (e.g., Scaffold-GS), demonstrating strong generalization capabilities.

Reconstructing and understanding 3D structures from a limited number of images is a classical problem in computer vision. Traditional approaches typically decompose this task into multiple subtasks, involving several stages of complex mappings between different data representations. For example, dense reconstruction using Structure-from-Motion (SfM) requires transforming images into key points, optimizing camera parameters, and estimating structures. Following this, accurate sparse reconstructions are necessary for further dense modeling, which is then input into task-specific neural networks. This multi-stage paradigm leads to significant processing times and engineering complexity. In this work, we introduce the Large Spatial Model (LSM), which directly processes unposed RGB images into semantic radiance fields. LSM simultaneously estimates geometry, appearance, and semantics in a single feed-forward pass and can synthesize versatile label maps by interacting through language at novel views. Built on a general Transformer-based framework, LSM predicts global geometry via pixel-aligned point maps. To improve spatial attribute regression, we adopt local context aggregation with multi-scale fusion, enhancing the accuracy of fine local details. To address the scarcity of labeled 3D semantic data and enable natural language-driven scene manipulation, we incorporate a pre-trained 2D language-based segmentation model into a 3D-consistent semantic feature field. An efficient decoder parameterizes a set of semantic anisotropic Gaussians, allowing supervised end-to-end learning. Comprehensive experiments on various tasks demonstrate that LSM unifies multiple 3D vision tasks directly from unposed images, achieving real-time semantic 3D reconstruction for the first time.

Nuanced expressiveness, especially through detailed hand and facial expressions, is pivotal for enhancing the realism and vitality of digital human representations. In this work, we aim to learn expressive human avatars from a monocular RGB video; a setting that introduces new challenges in capturing and animating fine-grained details. To this end, we introduce EVA, a drivable human model that can recover fine details based on 3D Gaussians and an expressive parametric human model, SMPL-X. Focused on enhancing expressiveness, our work makes three key contributions. First, we highlight the importance of aligning the SMPL-X model with the video frames for effective avatar learning. Recognizing the limitations of current methods for estimating SMPL-X parameters from in-the-wild videos, we introduce a reconstruction module that significantly improves the image-model alignment. Second, we propose a context-aware adaptive density control strategy, which is adaptively adjusting the gradient thresholds to accommodate the varied granularity across body parts. Third, we develop a feedback mechanism that predicts per-pixel confidence to better guide the optimization of 3D Gaussians. Extensive experiments on two benchmarks demonstrate the superiority of our approach both quantitatively and qualitatively, especially on the fine-grained hand and facial details. We make our code available at the project website: https://evahuman.github.io.

In this paper, we propose a data-model-hardware tri-design framework for high-throughput, low-cost, and high-accuracy multi-object tracking (MOT) on High-Definition (HD) video stream. First, to enable ultra-light video intelligence, we propose temporal frame-filtering and spatial saliency-focusing approaches to reduce the complexity of massive video data. Second, we exploit structure-aware weight sparsity to design a hardware-friendly model compression method. Third, assisted with data and model complexity reduction, we propose a sparsity-aware, scalable, and low-power accelerator design, aiming to deliver real-time performance with high energy efficiency. Different from existing works, we make a solid step towards the synergized software/hardware co-optimization for realistic MOT model implementation. Compared to the state-of-the-art MOT baseline, our tri-design approach can achieve 12.5× latency reduction, 20.9× effective frame rate improvement, 5.83× lower power, and 9.78× better energy efficiency, without much accuracy drop.

Virtual reality and augmented reality (XR) bring increasing demand for 3D content generation. However, creating high-quality 3D content requires tedious work from a human expert. In this work, we study the challenging task of lifting a single image to a 3D object and, for the first time, demonstrate the ability to generate a plausible 3D object with 360° views that corresponds well with the given reference image. By conditioning on the reference image, our model can fulfill the everlasting curiosity for synthesizing novel views of objects from images. Our technique sheds light on a promising direction of easing the workflows for 3D artists and XR designers. We propose a novel framework, dubbed NeuralLift-360, that utilizes a depth-aware neural radiance representation (NeRF) and learns to craft the scene guided by denoising diffusion models. By introducing a ranking loss, our NeuralLift-360 can be guided with rough depth estimation in the wild. We also adopt a CLIP-guided sampling strategy for the diffusion prior to provide coherent guidance. Extensive experiments demonstrate that our NeuralLift-360 significantly outperforms existing state-of-the-art baselines. Project page: https://vita-group.github.io/NeuralLift-360/

An increasing number of researchers are finding use for nth-order gradient computations for a wide variety of applications, including graphics, meta-learning (MAML), scientific computing, and most recently, implicit neural representations (INRs). Recent work shows that the gradient of an INR can be used to edit the data it represents directly without needing to convert it back to a discrete representation. However, given a function represented as a computation graph, traditional architectures face challenges in efficiently computing its nth-order gradient due to the higher demand for computing power and higher complexity in data movement. This makes it a promising target for FPGA acceleration. In this work, we introduce INR-Arch, a framework that transforms the computation graph of an nth-order gradient into a hardware-optimized dataflow architecture. We address this problem in two phases. First, we design a dataflow architecture that uses FIFO streams and an optimized computation kernel library, ensuring high memory efficiency and parallel computation. Second, we propose a compiler that extracts and optimizes computation graphs, automatically configures hardware parameters such as latency and stream depths to optimize throughput, while ensuring deadlock-free operation, and outputs High-Level Synthesis (HLS) code for FPGA implementation. We utilize INR editing as our benchmark, presenting results that demonstrate 1.8-4.8x and 1.5-3.6x speedup compared to CPU and GPU baselines respectively. Furthermore, we obtain 3.1-8.9x and 1.7-4.3x lower memory usage, and 1.7-11.3x and 5.5-32.8x lower energy-delay product. Our framework will be made open-source and available on GitHub.****https://github.com/sharc-lab/inr-arch

The computer vision community is embracing two promising learning paradigms: the Vision Transformer (ViT) and Multi-task Learning (MTL). ViT models show extraordinary performance over traditional convolution networks but are commonly recognized as computation-intensive, especially the self-attention with quadratic complexity. MTL uses one model to infer multiple tasks with better performance by enforcing shared representation among tasks, but a huge drawback is that, most MTL regimes require activation of the entire model even when only one or a few tasks are needed, causing significant computing waste. M3ViT is the latest multi-task Vi$T$model that introduces mixture-of-experts (MoE), where only a small portion of subnetworks (“experts”) are sparsely and dynamically activated based on the current task. M3Vi$T$achieves better accuracy and over 80% computation reduction and paves the way for efficient real-time MTL using ViT. Despite the algorithmic advantages of MTL, ViT, and even M3ViT, there are still many challenges for efficient deployment on FPGA. For instance, in general Transformer/ViT models, the self-attention is known as computational intensive and requires high bandwidth. In addition, softmax operations and the activation function GELU are extensively used, which unfortunately can consume more than half of the entire FPGA resource (LUTs). In the M3ViT model, the promising MoE mechanism for multi-task exposes new challenges for memory access overhead and also increases resource usage because of more layer types. To address these challenges in both general Transformer/ViT models and the state-of-the-art multi-task M3ViT with MoE, we propose Edge-MoE, the first end-to-end FPGA accelerator for multi-task ViT with a rich collection of architectural innovations. First, for general Transformer/ViT models, we propose (1) a novel reordering mechanism for self-attention, which reduces the bandwidth requirement from proportional to constant regardless of the target parallelism; (2) a fast single-pass softmax approximation; (3) an accurate and low-cost GELU approximation, which can significantly reduce the computation latency and resource usage; and (4) a unified and flexible computing unit that can be shared by almost all computational layers to maximally reduce resource usage. Second, for the advanced multi-task M3ViT with MoE, we propose a novel patch reordering method to completely eliminate any memory access overhead. Third, we deliver on-board implementation and measurement on Xilinx ZCU102 FPGA, with verified functionality and open-sourced hardware design, which achieves 2.24× and 4.90× better energy efficiency comparing with GPU (A6000) and CPU (Xeon 6226R), respectively. A real-time video demonstration of our accelerated multi-task ViT on an autonomous driving dataset is available in GitHub,11https://github.com/sharc-lab/Edge-MoE/raw/main/demo.mp4 together with our FPGA design using High-Level Synthesis, host code, FPGA bitstream, and on-board performance results.

Cross-scene generalizable NeRF models, which can directly synthesize novel views of unseen scenes, have become a new spotlight of the NeRF field. Several existing attempts rely on increasingly end-to-end "neuralized" architectures, i.e., replacing scene representation and/or rendering modules with performant neural networks such as transformers, and turning novel view synthesis into a feed-forward inference pipeline. While those feedforward "neuralized" architectures still do not fit diverse scenes well out of the box, we propose to bridge them with the powerful Mixture-of-Experts (MoE) idea from large language models (LLMs), which has demonstrated superior generalization ability by balancing between larger overall model capacity and flexible per-instance specialization. Starting from a recent generalizable NeRF architecture called GNT [52], we first demonstrate that MoE can be neatly plugged in to enhance the model. We further customize a shared permanent expert and a geometry-aware consistency loss to enforce cross-scene consistency and spatial smoothness respectively, which are essential for generalizable view synthesis. Our proposed model, dubbed GNT with Mixture-of-View-Experts (GNT-MOVE), has experimentally shown state-of-the-art results when transferring to unseen scenes, indicating remarkably better cross-scene generalization in both zero-shot and few-shot settings. Our codes are available at https://github.com/VITA-Group/GNT-MOVE.

Recent advancements in neural rendering have paved the way for a future marked by the widespread distribution of visual data through the sharing of Neural Radiance Field (NeRF) model weights. However, while established techniques exist for embedding ownership or copyright information within conventional visual data such as images and videos, the challenges posed by the emerging NeRF format have remained unaddressed. In this paper, we introduce StegaNeRF, an innovative approach for steganographic information embedding within NeRF renderings. We have meticulously developed an optimization framework that enables precise retrieval of hidden information from images generated by NeRF, while ensuring the original visual quality of the rendered images to remain intact. Through rigorous experimentation, we assess the efficacy of our methodology across various potential deployment scenarios. Furthermore, we delve into the insights gleaned from our analysis. StegaNeRF represents an initial foray into the intriguing realm of infusing NeRF renderings with customizable, imperceptible, and recoverable information, all while minimizing any discernible impact on the rendered images. For more details, please visit our project page: https://xggnet.github.io/StegaNeRF/

Neural Radiance Field (NeRF) regresses a neural param-eterized scene by differentially rendering multi-view images with ground-truth supervision. However, when interpolating novel views, NeRF often yields inconsistent and visually non-smooth geometric results, which we consider as a generalization gap between seen and unseen views. Recent advances in convolutional neural networks have demonstrated the promise of advanced robust data augmentations, either random or learned, in enhancing both in-distribution and out-of-distribution generalization. Inspired by that, we propose Augmented NeRF (Aug-NeRF), which for the first time brings the power of robust data augmentations into regular-izing the NeRF training. Particularly, our proposal learns to seamlessly blend worst-case perturbations into three distinct levels of the NeRF pipeline with physical grounds, including (1) the input coordinates, to simulate imprecise camera parameters at image capture; (2) intermediate features, to smoothen the intrinsic feature manifold; and (3) pre-rendering output, to account for the potential degra-dation factors in the multi-view image supervision. Extensive results demonstrate that Aug-NeRF effectively boosts NeRF performance in both novel view synthesis (up to 1.5dB PSNR gain) and underlying geometry reconstruction. Fur-thermore, thanks to the implicit smooth prior injected by the triple-level augmentations, Aug-NeRF can even recover scenes from heavily corrupted images, a highly challenging setting untackled before. Our codes are available in https://github.com/VITA-Group/Aug-NeRF.

Understanding 2D computer-aided design (CAD) drawings plays a crucial role for creating 3D prototypes in architecture, engineering and construction (AEC) industries. The task of automated panoptic symbol spotting, i.e., to spot and parse both countable object instances (windows, doors, tables, etc.) and uncountable stuff (wall, railing, etc.) from CAD drawings, has recently drawn interests from the computer vision community. Unfortunately, the highly irregular ordering and orientations set major roadblocks for this task. Existing methods, based on convolutional neural networks (CNNs) and/or graph neural networks (GNNs), regress instance bounding boxes in the pixel domain and then convert the predictions into symbols. In this paper, we present a novel framework named CAD Transformer, that can painlessly modify existing vision transformer (ViT) backbones to tackle the above limitations for the panoptic symbol spotting task. CADTransformer tokenizes directly from the set of graphical primitives in CAD drawings, and correspondingly optimizes line-grained semantic and instance symbol spotting altogether by a pair of prediction heads. The backbone is further enhanced with a few plug-and-play modifications, including a neighborhood aware self-attention, hierarchical feature aggregation, and graphic entity position encoding, to bake in the structure prior while optimizing the efficiency. Besides, a new data augmentation method, termed Random Layer, is proposed by the layer-wise separation and recombination of a CAD drawing. Overall, CADTransformer significantly boosts the previous state-of-the-art from 0.595 to 0.685 in the panoptic quality (PQ) metric, on the recently released FloorPlanCAD dataset. We further demonstrate that our model can spot symbols with irregular shapes and arbitrary orientations. Our codes are available in https://github.com/VITA-Group/CADTransformer.

Representing visual signals by coordinate-based deep fully-connected networks has been shown advantageous in fitting complex details and solving inverse problems than discrete grid-based representation. However, acquiring such a continuous Implicit Neural Representation (INR) requires tedious per-scene training on tons of signal measurements, which limits its practicality. In this paper, we present a generic INR framework that achieves both data and training efficiency by learning a Neural Implicit Dictionary (NID) from a data collection and representing INR as a functional combination of wavelets sampled from the dictionary. Our NID assembles a group of coordinate-based subnetworks which are tuned to span the desired function space. After training, one can instantly and robustly acquire an unseen scene representation by solving the coding coefficients. To parallelly optimize a large group of networks, we borrow the idea from Mixture-of-Expert (MoE) to design and train our network with a sparse gating mechanism. Our experiments show that, NID can improve reconstruction of 2D images or 3D scenes by 2 orders of magnitude faster with up to 98% less input data. We further demonstrate various applications of NID in image inpainting and occlusion removal, which are considered to be challenging with vanilla INR. Our codes are available in https://github.com/VITA-Group/Neural-Implicit-Dict.

Multi-task learning (MTL) encapsulates multiple learned tasks in a single model and often lets those tasks learn better jointly. Multi-tasking models have become successful and often essential for many sophisticated systems such as autonomous driving and indoor robots. However, when deploying MTL onto those real-world systems that are often resource-constrained or latency-sensitive, two prominent challenges arise: (i) during training, simultaneously optimizing all tasks is often difficult due to gradient conflicts across tasks, and the challenge is amplified when a growing number of tasks have to be squeezed into one compact model; (ii) at inference, current MTL regimes have to activate nearly the entire model even to just execute a single task. Yet most real systems demand only one or two tasks at each moment, while flexibly switching between tasks per need: therefore such “all tasks activated” inference is also highly inefficient and non-scalable in practice. In this paper, we present a model-accelerator co-design framework to enable efficient on-device MTL, that tackles both training and inference bottlenecks. Our framework, dubbed M³ViT, customizes mixture-of-experts (MoE) layers into a vision transformer (ViT) backbone for MTL, and sparsely activates task-specific experts during training, which effectively disentangles the parameter spaces to avoid different tasks’ training conflicts. Then at inference with any task of interest, the same design allows for activating only the task-corresponding sparse “expert” pathway, instead of the full model. Our new model design is further enhanced by hardware-level innovations, in particular, a novel computation reordering scheme tailored for memory-constrained MTL that achieves zero-overhead switching between tasks and can scale to any number of experts. Extensive experiments on PASCAL-Context and NYUD-v2 datasets at both software and hardware levels are conducted to demonstrate the effectiveness of the proposed design. When executing the practical scenario of single-task inference, M³ViT achieves higher accuracies than encoder-focused MTL methods, while significantly reducing 88% inference FLOPs. When implemented on a hardware platform of one Xilinx ZCU104 FPGA, our co-design framework reduces the memory requirement by 2.40×, while achieving energy efficiency (as the product of latency and power) up to 9.23× times higher than a comparable FPGA baseline.

Implicit Neural Representations (INRs) encoding continuous multi-media data via multi-layer perceptrons has shown undebatable promise in various computer vision tasks. Despite many successful applications, editing and processing an INR remains intractable as signals are represented by latent parameters of a neural network. Existing works manipulate such continuous representations via processing on their discretized instance, which breaks down the compactness and continuous nature of INR. In this work, we present a pilot study on the question: how to directly modify an INR without explicit decoding? We answer this question by proposing an implicit neural signal processing network, dubbed INSP-Net, via differential operators on INR. Our key insight is that spatial gradients of neural networks can be computed analytically and are invariant to translation, while mathematically we show that any continuous convolution filter can be uniformly approximated by a linear combination of high-order differential operators. With these two knobs, INSP-Net instantiates the signal processing operator as a weighted composition of computational graphs corresponding to the high-order derivatives of INRs, where the weighting parameters can be data-driven learned. Based on our proposed INSP-Net, we further build the first Convolutional Neural Network (CNN) that implicitly runs on INRs, named INSP-ConvNet. Our experiments validate the expressiveness of INSP-Net and INSP-ConvNet in fitting low-level image and geometry processing kernels (e.g. blurring, deblurring, denoising, inpainting, and smoothening) as well as for high-level tasks on implicit fields such as image classification.

Deep learning based 3D shape generation methods generally utilize latent features extracted from color images to encode the semantics of objects and guide the shape generation process. These color image semantics only implicitly encode 3D information, potentially limiting the accuracy of the generated shapes. In this paper we propose a multi-view mesh generation method which incorporates geometry information explicitly by using the features from intermediate depth representations of multi-view stereo and regularizing the 3D shapes against these depth images. First, our system predicts a coarse 3D volume from the color images by probabilistically merging voxel occupancy grids from the prediction of individual views. Then the depth images from multi-view stereo along with the rendered depth images of the coarse shape are used as a contrastive input whose features guide the refinement of the coarse shape through a series of graph convolution networks. Notably, we achieve superior results than state-of-the-art multi-view shape generation methods with 34% decrease in Chamfer distance to ground truth and 14% increase in F1-score on ShapeNet dataset.

Access to large and diverse computer-aided design (CAD) drawings is critical for developing symbol spotting algorithms. In this paper, we present FloorPlan-CAD, a large-scale real-world CAD drawing dataset containing over 10,000 floor plans, ranging from residential to commercial buildings. CAD drawings in the dataset are all represented as vector graphics, which enable us to provide line-grained annotations of 30 object categories. Equipped by such annotations, we introduce the task of panoptic symbol spotting, which requires to spot not only instances of countable things, but also the semantic of uncountable stuff. Aiming to solve this task, we propose a novel method by combining Graph Convolutional Networks (GCNs) with Convolutional Neural Networks (CNNs), which captures both non-Euclidean and Euclidean features and can be trained end-to-end. The proposed CNN-GCN method achieved state-of-the-art (SOTA) performance on the task of semantic symbol spotting, and help us build a baseline network for the panoptic symbol spotting task. Our contributions are three-fold: 1) to the best of our knowledge, the presented CAD drawing dataset is the first of its kind; 2) the panoptic symbol spotting task considers the spotting of both thing instances and stuff semantic as one recognition problem; and 3) we presented a baseline solution to the panoptic symbol spotting task based on a novel CNN-GCN method, which achieved SOTA performance on semantic symbol spotting. We believe that these contributions will boost research in related areas. The dataset and code is publicly available at https://floorplancad.github.io/.

The deep multi-view stereo (MVS) and stereo matching approaches generally construct 3D cost volumes to regularize and regress the output depth or disparity. These methods are limited when high-resolution outputs are needed since the memory and time costs grow cubically as the volume resolution increases. In this paper, we propose a both memory and time efficient cost volume formulation that is complementary to existing multi-view stereo and stereo matching approaches based on 3D cost volumes. First, the proposed cost volume is built upon a standard feature pyramid encoding geometry and context at gradually finer scales. Then, we can narrow the depth (or disparity) range of each stage by the depth (or disparity) map from the previous stage. With gradually higher cost volume resolution and adaptive adjustment of depth (or disparity) intervals, the output is recovered in a coarser to fine manner. We apply the cascade cost volume to the representative MVS-Net, and obtain a 35.6% improvement on DTU benchmark (1st place), with 50.6% and 59.3% reduction in GPU memory and run-time. It is also the state-of-the-art learning-based method on Tanks and Temples benchmark. The statistics of accuracy, run-time and GPU memory on other representative stereo CNNs also validate the effectiveness of our proposed method. Our source code is available at https://github.com/alibaba/cascade-stereo.

Compressed sensing (CS) theory can accelerate multi-contrast magnetic resonance imaging (MRI) by sampling fewer measurements within each contrast. However, conventional optimization-based reconstruction models suffer several limitations, including a strict assumption of shared sparse support, time-consuming optimization, and "shallow" models with difficulties in encoding the patterns contained in massive MRI data. In this paper, we propose the first deep learning model for multi-contrast CS-MRI reconstruction. We achieve information sharing through feature sharing units, which significantly reduces the number of model parameters. The feature sharing unit combines with a data fidelity unit to comprise an inference block, which are then cascaded with dense connections, allowing for efficient information transmission across different depths of the network. Experiments on various multi-contrast MRI datasets show that the proposed model outperforms both state-of-the-art single-contrast and multi-contrast MRI methods in accuracy and efficiency. We demonstrate that improved reconstruction quality can bring benefits to subsequent medical image analysis. Furthermore, the robustness of the proposed model to misregistration shows its potential in real MRI applications.

Compressed sensing magnetic resonance imaging (CS-MRI) is an active research topic in the field of inverse problems. Conventional CS-MRI algorithms usually exploit the sparse nature of MRI in an iterative manner. These optimization-based CS-MRI methods are often time-consuming at test time, and are based on fixed transform bases or shallow dictionaries, which limits modeling capacity. Recently, deep models have been introduced to the CS-MRI problem. One main challenge for CS-MRI methods based on deep learning is the trade off between model performance and network size. We propose a recursive dilated network (RDN) for CS-MRI that achieves good performance while reducing the number of network parameters. We adopt dilated convolutions in each recursive block to aggregate multi-scale information within the MRI. We also adopt a modified shortcut strategy to help features flow into deeper layers. Experimental results show that the proposed RDN model achieves state-of-the-art performance in CS-MRI while using far fewer parameters than previously required.

Single image rain streaks removal is extremely important since rainy condition adversely affects many computer vision systems. Deep learning based methods have great success in image deraining tasks. In this paper, we propose a novel residual-guide feature fusion network, called ResGuideNet, for single image deraining that progressively predicts high-quality reconstruction while using fewer parameters than previous methods. Specifically, we propose a cascaded network and adopt residuals from shallower blocks to guide deeper blocks. We can obtain a coarse-to-fine estimation of negative residual as the blocks go deeper with this strategy. The outputs of different blocks are merged into the final reconstruction. We adopt recursive convolution to build each block and apply supervision to intermediate de-rained results. ResGuideNet is detachable to meet different rainy conditions. For images with light rain streaks and limited computational resource at test time, we can obtain a decent performance even with several building blocks. Experiments validate that ResGuideNet can benefit other low- and high-level vision tasks.