Imitation learning for robotic manipulation faces a fundamental challenge: the scarcity of large-scale, high-quality robot demonstration data. Recent robotic foundation models often pre-train on cross-embodiment robot datasets to increase data scale, while they face significant limitations as the diverse morphologies and action spaces across different robot embodiments make unified training challenging. In this paper, we present H-RDT (Human to Robotics Diffusion Transformer), a novel approach that leverages human manipulation data to enhance robot manipulation capabilities. Our key insight is that large-scale egocentric human manipulation videos with paired 3D hand pose annotations provide rich behavioral priors that capture natural manipulation strategies and can benefit robotic policy learning. We introduce a two-stage training paradigm: (1) pre-training on large-scale egocentric human manipulation data, and (2) cross-embodiment fine-tuning on robot-specific data with modular action encoders and decoders. Built on a diffusion transformer architecture with 2B parameters, H-RDT uses flow matching to model complex action distributions. The modular design of action encoder and decoder components enables effective knowledge transfer from the unified human embodiment to diverse robot platforms through efficient fine-tuning. Extensive evaluations encompassing both simulation and real-world experiments, single-task and multitask scenarios, as well as few-shot learning and robustness assessments, demonstrate that H-RDT outperforms training from scratch and existing state-of-the-art methods, including π0 and RDT, achieving significant improvements of 13.9% and 40.5% over training from scratch in simulation and real-world experiments, respectively. The results validate our core hypothesis that human manipulation data can serve as a powerful foundation for learning bimanual robotic manipulation policies.

Reconstructing complete and interactive 3D scenes remains a fundamental challenge in computer vision and robotics, particularly due to persistent object occlusions and limited sensor coverage. Even multi-view observations from a single scene scan often fail to capture the full structural details. Existing approaches typically rely on multi-stage pipelines—such as segmentation, background completion, and inpainting—or require per-object dense scanning, both of which are error-prone, and not easily scalable. We propose IGFuse, a novel framework that reconstructs interactive Gaussian scene by fusing observations from multiple scans, where natural object rearrangement between captures reveal previously occluded regions. Our method constructs segmentation-aware Gaussian fields and enforces bi-directional photometric and semantic consistency across scans. To handle spatial misalignments, we introduce a pseudo-intermediate scene state for symmetric alignment, alongside collaborative co-pruning strategies to refine geometry. IGFuse enables high-fidelity rendering and object-level scene manipulation without dense observations or complex pipelines. Extensive experiments validate the framework’s strong generalization to novel scene configurations, demonstrating its effectiveness for real-world 3D reconstruction and real-to-simulation transfer.

Vision-Language Navigation (VLN) tasks often leverage panoramic RGB and depth inputs to provide rich spatial cues for action planning, but these sensors can be costly or less accessible in real-world deployments. Recent approaches based on Vision-Language Action (VLA) models achieve strong results with monocular input, yet they still lag behind methods using panoramic RGB-D information. We present MonoDream, a lightweight VLA framework that enables monocular agents to learn a Unified Navigation Representation (UNR). This shared feature representation jointly aligns navigation-relevant visual semantics (e.g., global layout, depth, and future cues) and language-grounded action intent, enabling more reliable action prediction. MonoDream further introduces Latent Panoramic Dreaming (LPD) tasks to supervise the UNR, which train the model to predict latent features of panoramic RGB and depth observations at both current and future steps based on only monocular input. Experiments on multiple VLN benchmarks show that MonoDream consistently improves monocular navigation performance and significantly narrows the gap with panoramic-based agents.

In the field of autonomous driving, the structural design of perception models is of paramount importance. Unlike mainstream BEV-based algorithms, we propose a novel end-to-end sparse perception framework, Sparse4D, to achieve better performance and higher efficiency. Starting from the sparsity of perception results, we define an instance that decouples implicit features and explicit anchors, using the instance as the core for feature fusion to accomplish perception tasks. For spatial modelling, we develop a novel operator called deformable aggregation, which enables the transfer of information from the dense image feature to the sparse instance. For temporal modelling, we design a recurrent instance feature propagation structure, which not only realizes long-term feature fusion but also ensures the computational efficiency of the temporal module. Lastly, we explored the performance of Sparse4D in multi-object tracking tasks and proposed a minimalist joint detection and tracking model. We conduct extensive experimental validation of Sparse4D on the nuScenes benchmark. Sparse4D achieved state-of-the-art performance in multi-camera 3D detection and tracking tasks, and it also outperformed other algorithms in terms of training and inference efficiency. Furthermore, we extended Sparse4D to a multi-modal model, achieving excellent performance and better generalization.

The labor- and experience-intensive creation of 3D assets with physically based rendering (PBR) materials demands an autonomous 3D asset creation pipeline. However, most existing 3D generation methods focus on geometry modeling, either baking textures into simple vertex colors or leaving texture synthesis to post-processing with image diffusion models. To achieve end-to-end PBR-ready 3D asset generation, we present Lightweight Gaussian Asset Adapter (LGAA), a novel framework that unifies the modeling of geometry and PBR materials by exploiting multi-view (MV) diffusion priors from a novel perspective. The LGAA features a modular design with three components. Specifically, the LGAA Wrapper reuses and adapts network layers from MV diffusion models, which encapsulate knowledge acquired from billions of images, enabling better convergence in a data-efficient manner. To incorporate multiple diffusion priors for geometry and PBR synthesis, the LGAA Switcher aligns multiple LGAA Wrapper layers encapsulating different knowledge. Then, a tamed variational autoencoder (VAE), termed LGAA Decoder, is designed to predict 2D Gaussian Splatting (2DGS) with PBR channels. Finally, we introduce a dedicated post-processing procedure to effectively extract high-quality, relightable mesh assets from the resulting 2DGS. Extensive quantitative and qualitative experiments demonstrate the superior performance of LGAA with both text- and image-conditioned MV diffusion models. Additionally, the modular design enables flexible incorporation of multiple diffusion priors, and the knowledge-preserving scheme effectively preseves the 2D priors learned on massive image dataset, which leads to data efficient finetuning to lift the MV diffuison models for 3D generation with merely 69k multi-view instances. Our code, pre-trained weights, and the dataset used will be publicly available via our project page: https://zx-yin.github.io/dreamlifting/.

3D Semantic Occupancy Prediction is fundamental for spatial understanding, yet existing approaches face challenges in scalability and generalization due to their reliance on extensive labeled data and computationally intensive voxel-wise representations. In this paper, we introduce GaussTR, a novel Gaussian-based TRansformer framework that unifies sparse 3D modeling with foundation model alignment through Gaussian representations to advance 3D spatial understanding. GaussTR predicts sparse sets of Gaussians in a feed-forward manner to represent 3D scenes. By splatting the Gaussians into 2D views and aligning the rendered features with foundation models, GaussTR facilitates self-supervised 3D representation learning and enables open-vocabulary semantic occupancy prediction without requiring explicit annotations. Empirical experiments on the Occ3D-nuScenes dataset demonstrate GaussTR’s state-of-the-art zero-shot performance of 12.27 mIoU, along with a 40% reduction in training time. These results highlight the efficacy of GaussTR for scalable and holistic 3D spatial understanding, with promising implications in autonomous driving and embodied agents. The code is available at https://github.com/hustvl/GaussTR.

In embodied intelligence systems, a key component is 3D perception algorithm, which enables agents to understand their surrounding environments. Previous algorithms primarily rely on point cloud, which, despite offering precise geometric information, still constrain perception performance due to inherent sparsity, noise, and data scarcity. In this work, we introduce a novel image-centric 3D perception model, BIP3D, which leverages expressive image features with explicit 3D position encoding to overcome the limitations of point-centric methods. Specifically, we leverage pre-trained 2D vision foundation models to enhance semantic understanding, and introduce a spatial enhancer module to improve spatial understanding. Together, these modules enable BIP3D to achieve multi-view, multi-modal feature fusion and end-to-end 3D perception. In our experiments, BIP3D outperforms current state-of-the-art results on the EmbodiedScan benchmark, achieving improvements of 5.69% in the 3D detection task and 15.25% in the 3D visual grounding task. Code has been released at https://github.com/HorizonRobotics/BIP3D.

This paper presents GeoFlow-SLAM, a robust and effective Tightly-Coupled RGBD-Inertial and Legged Odometry Fusion SLAM for legged robotics undergoing aggressive and high-frequency motions. By integrating geometric consistency, legged odometry constraints, and dual-stream optical flow (GeoFlow), our method addresses three critical challenges: feature matching and pose initialization failures during fast locomotion and visual feature scarcity in texture-less scenes. Specifically, in rapid motion scenarios, feature matching is notably enhanced by leveraging dual-stream optical flow, which combines prior map points and poses. Additionally, we propose a robust pose initialization method for fast locomotion and IMU error in legged robots, integrating IMU/Legged odometry, inter-frame Perspective-n-Point (PnP), and Generalized Iterative Closest Point (GICP). Furthermore, a novel optimization framework that tightly couples depth-to-map and GICP geometric constraints is first introduced to improve the robustness and accuracy in long-duration, visually texture-less environments. The proposed algorithms achieve state-of-the-art (SOTA) on collected legged robots and open-source datasets. To further promote research and development, the open-source datasets and code will be made publicly available at https://github.com/HorizonRobotics/GeoFlowSlam.

Vision-Language Navigation is a critical task for developing embodied agents that can follow natural language instructions to navigate in complex real-world environments. Recent advances by finetuning large pretrained models have significantly improved generalization and instruction grounding compared to traditional approaches. However, the role of reasoning strategies in navigation—an action-centric, long-horizon task—remains underexplored, despite Chain-of-Thought reasoning's demonstrated success in static tasks like question answering and visual reasoning. To address this gap, we conduct the first systematic evaluation of reasoning strategies for VLN, including No-Think (direct action prediction), Pre-Think (reason before action), and Post-Think (reason after action). Surprisingly, our findings reveal the Inference-time Reasoning Collaps issue, where inference-time reasoning degrades navigation accuracy, highlighting the challenges of integrating reasoning into VLN. Based on this insight, we propose Aux-Think, a framework that trains models to internalize structured reasoning patterns through CoT supervision during training, while preserving No-Think inference for efficient action prediction. To support this framework, we release R2R-CoT-320k, a large-scale Chain-of-Thought annotated dataset. Empirically, Aux-Think significantly reduces training effort without compromising performance.

We present **DIPO**, a novel framework for the controllable generation of articulated 3D objects from a pair of images: one depicting the object in a resting state and the other in an articulated state. Compared to the single-image approach, our dual-image input imposes only a modest overhead for data collection, but at the same time provides important motion information, which is a reliable guide for predicting kinematic relationships between parts. Specifically, we propose a dual-image diffusion model that captures relationships between the image pair to generate part layouts and joint parameters. In addition, we introduce a Chain-of-Thought (CoT) based **graph reasoner** that explicitly infers part connectivity relationships. To further improve robustness and generalization on complex articulated objects, we develop a fully automated dataset expansion pipeline, name **LEGO-Art**, that enriches the diversity and complexity of PartNet-Mobility dataset. We propose **PM-X**, a large-scale dataset of complex articulated 3D objects, accompanied by rendered images, URDF annotations, and textual descriptions. Extensive experiments demonstrate that DIPO significantly outperforms existing baselines in both the resting state and the articulated state, while the proposed PM-X dataset further enhances generalization to diverse and structurally complex articulated objects. Our code and dataset are available at https://github.com/RQ-Wu/DIPO.

In this work, we present a new conceptually simple and flexible network for accurate scene text extraction, which handle text detection and text recognition concurrently. Apart from solving feature sharing by designing a unified network, we implement end-to-end training of the whole system by a set of new optimization strategies. More importantly, this method highlights its novel parallel detection-recognition structure, which constructs a loose connection between both detection and recognition. This loose connection is embodied in the definition of overall loss and derivative back propagation for the model parameter updating, which automatically balances the contribution of the two branches to the system performance. It is different from the existing end-to-end methods where two subtasks are connected serially and thus yielding heavy dependence of the predecessor text detection task on the follow-up text recognition task and sensitivity of recognition to detection noise. In addition, a simple Mask-Rectifier mechanism is applied to easily adapt our system to incidental text recognition with arbitrary orientation and shape. Experiment results on Incidental Scene Text ICDAR2015 dataset surpass the current state-of-the-art FOTS method, as suggests the effectiveness of the proposed approach.

Pixel-level labelling tasks, such as semantic segmentation, play a central role in image understanding. Recent approaches have attempted to harness the capabilities of deep learning techniques for image recognition to tackle pixel-level labelling tasks. One central issue in this methodology is the limited capacity of deep learning techniques to delineate visual objects. To solve this problem, we introduce a new form of convolutional neural network that combines the strengths of Convolutional Neural Networks (CNNs) and Conditional Random Fields (CRFs)-based probabilistic graphical modelling. To this end, we formulate Conditional Random Fields with Gaussian pairwise potentials and mean-field approximate inference as Recurrent Neural Networks. This network, called CRF-RNN, is then plugged in as a part of a CNN to obtain a deep network that has desirable properties of both CNNs and CRFs. Importantly, our system fully integrates CRF modelling with CNNs, making it possible to train the whole deep network end-to-end with the usual back-propagation algorithm, avoiding offline post-processing methods for object delineation. We apply the proposed method to the problem of semantic image segmentation, obtaining top results on the challenging Pascal VOC 2012 segmentation benchmark.