arXiv subjects

Sanping Zhou

Publications and source records attributed to Sanping Zhou.

At least 19 recordsLinked to original sources

arXiv2026

GE-Act 2.0: Pretraining and Scaling a World-Action Model for Robotic Manipulation

World-action models (WAM) predict future states to guide robot actions, enabling learning from both action-free video and action-labeled interaction. Most inherit pretrained video generators, leaving WAM pretraining and scaling underexplored. We introduce Genie Envisioner Act 2.0 (GE-Act 2.0), a world-action model whose trainable generative and action components are all initialized from scratch on manipulation data. It combines a control-oriented autoencoder (CoAE), a single-step visual planner (SVP), and an inverse dynamics model (IDM). CoAE retains action- and instruction-relevant information under aggressive compression, while SVP produces a complete future state in one differentiable pass, so visual planning and inverse dynamics can be pretrained separately on complementary data. The components are then jointly trained with knowledge-aligned selective optimization (KASO), which reduces mismatched supervision by selecting only predicted futures judged behaviorally compatible with the recorded action. We evaluate pretrained checkpoints directly, without per-task fine-tuning, on 100 tasks across 20 manipulation skill groups with held-out scenes, backgrounds, lighting, and object instances. Scaling co-training data from 300 to 30,000 hours raises success from 17.1% to 44.1% on G1-OP and from 13.4% to 31.1% on G2-90D; despite comprising less than 2% of the co-training data, G2-90D improves by 17.7 points, suggesting cross-embodiment transfer. Gains span 19/20 and 18/20 skill groups, and skill-specific coverage strongly correlates with zero-shot out-of-distribution (OOD) success (Pearson r=0.80; Spearman rho=0.85). Under the same protocol, the model grounds object, color, shape, and position references in at least 90% of trials and follows explicit instructions even when they conflict with an already-committed behavior or a conventional scene association.

cs.RO↗

arXiv2026

Efficient Camera Pose Augmentation for View Generalization in Robotic Policy Learning

Prevailing 2D-centric visuomotor policies exhibit a pronounced deficiency in novel view generalization, as their reliance on static observations hinders consistent action mapping across unseen views. In response, we introduce GenSplat, a feed-forward 3D Gaussian Splatting framework that facilitates view-generalized policy learning through novel view rendering. GenSplat employs a permutation-equivariant architecture to reconstruct high-fidelity 3D scenes from sparse, uncalibrated inputs in a single forward pass. To ensure structural integrity, we design a 3D-prior distillation strategy that regularizes the 3DGS optimization, preventing the geometric collapse typical of purely photometric supervision. By rendering diverse synthetic views from these stable 3D representations, we systematically augment the observational manifold during training. This augmentation forces the policy to ground its decisions in underlying 3D structures, thereby ensuring robust execution under severe spatial perturbations where baselines severely degrade.

cs.RO↗

arXiv2026

Robust Self-Training with Closed-loop Label Correction for Learning from Noisy Labels

Training deep neural networks with noisy labels remains a significant challenge, often leading to degraded performance. Existing methods for handling label noise typically rely on either transition matrix, noise detection, or meta-learning techniques, but they often exhibit low utilization efficiency of noisy samples and incur high computational costs. In this paper, we propose a self-training label correction framework using decoupled bilevel optimization, where a classifier and neural correction function co-evolve. Leveraging a small clean dataset, our method employs noisy posterior simulation and intermediate features to transfer ground-truth knowledge, forming a closed-loop feedback system that prevents error amplification. Theoretical guarantees underpin the stability of our approach, and extensive experiments on benchmark datasets like CIFAR and Clothing1M confirm state-of-the-art performance with reduced training time, highlighting its practical applicability for learning from noisy labels.

cs.LG↗

arXiv2025

UniLayDiff: A Unified Diffusion Transformer for Content-Aware Layout Generation

Content-aware layout generation is a critical task in graphic design automation, focused on creating visually appealing arrangements of elements that seamlessly blend with a given background image. The variety of real-world applications makes it highly challenging to develop a single model capable of unifying the diverse range of input-constrained generation sub-tasks, such as those conditioned by element types, sizes, or their relationships. Current methods either address only a subset of these tasks or necessitate separate model parameters for different conditions, failing to offer a truly unified solution. In this paper, we propose UniLayDiff: a Unified Diffusion Transformer, that for the first time, addresses various content-aware layout generation tasks with a single, end-to-end trainable model. Specifically, we treat layout constraints as a distinct modality and employ Multi-Modal Diffusion Transformer framework to capture the complex interplay between the background image, layout elements, and diverse constraints. Moreover, we integrate relation constraints through fine-tuning the model with LoRA after pretraining the model on other tasks. Such a schema not only achieves unified conditional generation but also enhances overall layout quality. Extensive experiments demonstrate that UniLayDiff achieves state-of-the-art performance across from unconditional to various conditional generation tasks and, to the best of our knowledge, is the first model to unify the full range of content-aware layout generation tasks.

cs.CV↗

arXiv2025

FreqGRL: Suppressing Low-Frequency Bias and Mining High-Frequency Knowledge for Cross-Domain Few-Shot Learning

Cross-domain few-shot learning (CD-FSL) aims to recognize novel classes with only a few labeled examples under significant domain shifts. While recent approaches leverage a limited amount of labeled target-domain data to improve performance, the severe imbalance between abundant source data and scarce target data remains a critical challenge for effective representation learning. We present the first frequency-space perspective to analyze this issue and identify two key challenges: (1) models are easily biased toward source-specific knowledge encoded in the low-frequency components of source data, and (2) the sparsity of target data hinders the learning of high-frequency, domain-generalizable features. To address these challenges, we propose \textbf{FreqGRL}, a novel CD-FSL framework that mitigates the impact of data imbalance in the frequency space. Specifically, we introduce a Low-Frequency Replacement (LFR) module that substitutes the low-frequency components of source tasks with those from the target domain to create new source tasks that better align with target characteristics, thus reducing source-specific biases and promoting generalizable representation learning. We further design a High-Frequency Enhancement (HFE) module that filters out low-frequency components and performs learning directly on high-frequency features in the frequency space to improve cross-domain generalization. Additionally, a Global Frequency Filter (GFF) is incorporated to suppress noisy or irrelevant frequencies and emphasize informative ones, mitigating overfitting risks under limited target supervision. Extensive experiments on five standard CD-FSL benchmarks demonstrate that our frequency-guided framework achieves state-of-the-art performance.

cs.CV↗

arXiv2025

DynaRend: Learning 3D Dynamics via Masked Future Rendering for Robotic Manipulation

Learning generalizable robotic manipulation policies remains a key challenge due to the scarcity of diverse real-world training data. While recent approaches have attempted to mitigate this through self-supervised representation learning, most either rely on 2D vision pretraining paradigms such as masked image modeling, which primarily focus on static semantics or scene geometry, or utilize large-scale video prediction models that emphasize 2D dynamics, thus failing to jointly learn the geometry, semantics, and dynamics required for effective manipulation. In this paper, we present DynaRend, a representation learning framework that learns 3D-aware and dynamics-informed triplane features via masked reconstruction and future prediction using differentiable volumetric rendering. By pretraining on multi-view RGB-D video data, DynaRend jointly captures spatial geometry, future dynamics, and task semantics in a unified triplane representation. The learned representations can be effectively transferred to downstream robotic manipulation tasks via action value map prediction. We evaluate DynaRend on two challenging benchmarks, RLBench and Colosseum, as well as in real-world robotic experiments, demonstrating substantial improvements in policy success rate, generalization to environmental perturbations, and real-world applicability across diverse manipulation tasks.

cs.RO↗

arXiv2025

DAMap: Distance-aware MapNet for High Quality HD Map Construction

Predicting High-definition (HD) map elements with high quality (high classification and localization scores) is crucial to the safety of autonomous driving vehicles. However, current methods perform poorly in high quality predictions due to inherent task misalignment. Two main factors are responsible for misalignment: 1) inappropriate task labels due to one-to-many matching queries sharing the same labels, and 2) sub-optimal task features due to task-shared sampling mechanism. In this paper, we reveal two inherent defects in current methods and develop a novel HD map construction method named DAMap to address these problems. Specifically, DAMap consists of three components: Distance-aware Focal Loss (DAFL), Hybrid Loss Scheme (HLS), and Task Modulated Deformable Attention (TMDA). The DAFL is introduced to assign appropriate classification labels for one-to-many matching samples. The TMDA is proposed to obtain discriminative task-specific features. Furthermore, the HLS is proposed to better utilize the advantages of the DAFL. We perform extensive experiments and consistently achieve performance improvement on the NuScenes and Argoverse2 benchmarks under different metrics, baselines, splits, backbones, and schedules. Code will be available at https://github.com/jpdong-xjtu/DAMap.

cs.CV↗

arXiv2025

FreqPDE: Rethinking Positional Depth Embedding for Multi-View 3D Object Detection Transformers

Detecting 3D objects accurately from multi-view 2D images is a challenging yet essential task in the field of autonomous driving. Current methods resort to integrating depth prediction to recover the spatial information for object query decoding, which necessitates explicit supervision from LiDAR points during the training phase. However, the predicted depth quality is still unsatisfactory such as depth discontinuity of object boundaries and indistinction of small objects, which are mainly caused by the sparse supervision of projected points and the use of high-level image features for depth prediction. Besides, cross-view consistency and scale invariance are also overlooked in previous methods. In this paper, we introduce Frequency-aware Positional Depth Embedding (FreqPDE) to equip 2D image features with spatial information for 3D detection transformer decoder, which can be obtained through three main modules. Specifically, the Frequency-aware Spatial Pyramid Encoder (FSPE) constructs a feature pyramid by combining high-frequency edge clues and low-frequency semantics from different levels respectively. Then the Cross-view Scale-invariant Depth Predictor (CSDP) estimates the pixel-level depth distribution with cross-view and efficient channel attention mechanism. Finally, the Positional Depth Encoder (PDE) combines the 2D image features and 3D position embeddings to generate the 3D depth-aware features for query decoding. Additionally, hybrid depth supervision is adopted for complementary depth learning from both metric and distribution aspects. Extensive experiments conducted on the nuScenes dataset demonstrate the effectiveness and superiority of our proposed method.

cs.CV↗

arXiv2025

SAMPO:Scale-wise Autoregression with Motion PrOmpt for generative world models

World models allow agents to simulate the consequences of actions in imagined environments for planning, control, and long-horizon decision-making. However, existing autoregressive world models struggle with visually coherent predictions due to disrupted spatial structure, inefficient decoding, and inadequate motion modeling. In response, we propose \textbf{S}cale-wise \textbf{A}utoregression with \textbf{M}otion \textbf{P}r\textbf{O}mpt (\textbf{SAMPO}), a hybrid framework that combines visual autoregressive modeling for intra-frame generation with causal modeling for next-frame generation. Specifically, SAMPO integrates temporal causal decoding with bidirectional spatial attention, which preserves spatial locality and supports parallel decoding within each scale. This design significantly enhances both temporal consistency and rollout efficiency. To further improve dynamic scene understanding, we devise an asymmetric multi-scale tokenizer that preserves spatial details in observed frames and extracts compact dynamic representations for future frames, optimizing both memory usage and model performance. Additionally, we introduce a trajectory-aware motion prompt module that injects spatiotemporal cues about object and robot trajectories, focusing attention on dynamic regions and improving temporal consistency and physical realism. Extensive experiments show that SAMPO achieves competitive performance in action-conditioned video prediction and model-based control, improving generation quality with 4.4$\times$ faster inference. We also evaluate SAMPO's zero-shot generalization and scaling behavior, demonstrating its ability to generalize to unseen tasks and benefit from larger model sizes.

cs.CV↗

arXiv2025

Action Hints: Semantic Typicality and Context Uniqueness for Generalizable Skeleton-based Video Anomaly Detection

Zero-Shot Video Anomaly Detection (ZS-VAD) requires temporally localizing anomalies without target domain training data, which is a crucial task due to various practical concerns, e.g., data privacy or new surveillance deployments. Skeleton-based approach has inherent generalizable advantages in achieving ZS-VAD as it eliminates domain disparities both in background and human appearance. However, existing methods only learn low-level skeleton representation and rely on the domain-limited normality boundary, which cannot generalize well to new scenes with different normal and abnormal behavior patterns. In this paper, we propose a novel zero-shot video anomaly detection framework, unlocking the potential of skeleton data via action typicality and uniqueness learning. Firstly, we introduce a language-guided semantic typicality modeling module that projects skeleton snippets into action semantic space and distills LLM's knowledge of typical normal and abnormal behaviors during training. Secondly, we propose a test-time context uniqueness analysis module to finely analyze the spatio-temporal differences between skeleton snippets and then derive scene-adaptive boundaries. Without using any training samples from the target domain, our method achieves state-of-the-art results against skeleton-based methods on four large-scale VAD datasets: ShanghaiTech, UBnormal, NWPU, and UCF-Crime, featuring over 100 unseen surveillance scenes.

cs.CV↗

arXiv2025

Enhancing Mamba Decoder with Bidirectional Interaction in Multi-Task Dense Prediction

Sufficient cross-task interaction is crucial for success in multi-task dense prediction. However, sufficient interaction often results in high computational complexity, forcing existing methods to face the trade-off between interaction completeness and computational efficiency. To address this limitation, this work proposes a Bidirectional Interaction Mamba (BIM), which incorporates novel scanning mechanisms to adapt the Mamba modeling approach for multi-task dense prediction. On the one hand, we introduce a novel Bidirectional Interaction Scan (BI-Scan) mechanism, which constructs task-specific representations as bidirectional sequences during interaction. By integrating task-first and position-first scanning modes within a unified linear complexity architecture, BI-Scan efficiently preserves critical cross-task information. On the other hand, we employ a Multi-Scale Scan~(MS-Scan) mechanism to achieve multi-granularity scene modeling. This design not only meets the diverse granularity requirements of various tasks but also enhances nuanced cross-task feature interactions. Extensive experiments on two challenging benchmarks, \emph{i.e.}, NYUD-V2 and PASCAL-Context, show the superiority of our BIM vs its state-of-the-art competitors.

cs.CV↗

arXiv2025

Embracing Aleatoric Uncertainty: Generating Diverse 3D Human Motion

Generating 3D human motions from text is a challenging yet valuable task. The key aspects of this task are ensuring text-motion consistency and achieving generation diversity. Although recent advancements have enabled the generation of precise and high-quality human motions from text, achieving diversity in the generated motions remains a significant challenge. In this paper, we aim to overcome the above challenge by designing a simple yet effective text-to-motion generation method, \textit{i.e.}, Diverse-T2M. Our method introduces uncertainty into the generation process, enabling the generation of highly diverse motions while preserving the semantic consistency of the text. Specifically, we propose a novel perspective that utilizes noise signals as carriers of diversity information in transformer-based methods, facilitating a explicit modeling of uncertainty. Moreover, we construct a latent space where text is projected into a continuous representation, instead of a rigid one-to-one mapping, and integrate a latent space sampler to introduce stochastic sampling into the generation process, thereby enhancing the diversity and uncertainty of the outputs. Our results on text-to-motion generation benchmark datasets~(HumanML3D and KIT-ML) demonstrate that our method significantly enhances diversity while maintaining state-of-the-art performance in text consistency.

cs.CV↗

arXiv2025

CEM-Net: Cross-Emotion Memory Network for Emotional Talking Face Generation

Emotional talking face generation aims to animate a human face in given reference images and generate a talking video that matches the content and emotion of driving audio. However, existing methods neglect that reference images may have a strong emotion that conflicts with the audio emotion, leading to severe emotion inaccuracy and distorted generated results. To tackle the issue, we introduce a cross-emotion memory network(CEM-Net), designed to generate emotional talking faces aligned with the driving audio when reference images exhibit strong emotion. Specifically, an Audio Emotion Enhancement module(AEE) is first devised with the cross-reconstruction training strategy to enhance audio emotion, overcoming the disruption from reference image emotion. Secondly, since reference images cannot provide sufficient facial motion information of the speaker under audio emotion, an Emotion Bridging Memory module(EBM) is utilized to compensate for the lacked information. It brings in expression displacement from the reference image emotion to the audio emotion and stores it in the memory.Given a cross-emotion feature as a query, the matching displacement can be retrieved at inference time. Extensive experiments have demonstrated that our CEM-Net can synthesize expressive, natural and lip-synced talking face videos with better emotion accuracy.

cs.MM↗

arXiv2025

TSPO: Temporal Sampling Policy Optimization for Long-form Video Language Understanding

Multimodal Large Language Models (MLLMs) have demonstrated significant progress in vision-language tasks, yet they still face challenges when processing long-duration video inputs. The limitation arises from MLLMs' context limit and training costs, necessitating sparse frame sampling before feeding videos into MLLMs. However, building a trainable sampling method remains challenging due to the unsupervised and non-differentiable nature of sparse frame sampling in Video-MLLMs. To address these problems, we propose Temporal Sampling Policy Optimization (TSPO), advancing MLLMs' long-form video-language understanding via reinforcement learning. Specifically, we first propose a trainable event-aware temporal agent, which captures event-query correlation for performing probabilistic keyframe selection. Then, we propose the TSPO reinforcement learning paradigm, which models keyframe selection and language generation as a joint decision-making process, enabling end-to-end group relative optimization for the temporal sampling policy. Furthermore, we propose a dual-style long video training data construction pipeline, balancing comprehensive temporal understanding and key segment localization. Finally, we incorporate rule-based answering accuracy and temporal locating reward mechanisms to optimize the temporal sampling policy. Comprehensive experiments show that our TSPO achieves state-of-the-art performance across multiple long video understanding benchmarks, and shows transferable ability across different cutting-edge Video-MLLMs. Our code is available at https://github.com/Hui-design/TSPO

cs.CV↗

arXiv2025

Mind the Gap: Aligning Vision Foundation Models to Image Feature Matching

Leveraging the vision foundation models has emerged as a mainstream paradigm that improves the performance of image feature matching. However, previous works have ignored the misalignment when introducing the foundation models into feature matching. The misalignment arises from the discrepancy between the foundation models focusing on single-image understanding and the cross-image understanding requirement of feature matching. Specifically, 1) the embeddings derived from commonly used foundation models exhibit discrepancies with the optimal embeddings required for feature matching; 2) lacking an effective mechanism to leverage the single-image understanding ability into cross-image understanding. A significant consequence of the misalignment is they struggle when addressing multi-instance feature matching problems. To address this, we introduce a simple but effective framework, called IMD (Image feature Matching with a pre-trained Diffusion model) with two parts: 1) Unlike the dominant solutions employing contrastive-learning based foundation models that emphasize global semantics, we integrate the generative-based diffusion models to effectively capture instance-level details. 2) We leverage the prompt mechanism in generative model as a natural tunnel, propose a novel cross-image interaction prompting module to facilitate bidirectional information interaction between image pairs. To more accurately measure the misalignment, we propose a new benchmark called IMIM, which focuses on multi-instance scenarios. Our proposed IMD establishes a new state-of-the-art in commonly evaluated benchmarks, and the superior improvement 12% in IMIM indicates our method efficiently mitigates the misalignment.

cs.CV↗

arXiv2025

PR-DETR: Injecting Position and Relation Prior for Dense Video Captioning

Dense video captioning is a challenging task that aims to localize and caption multiple events in an untrimmed video. Recent studies mainly follow the transformer-based architecture to jointly perform the two sub-tasks, i.e., event localization and caption generation, in an end-to-end manner. Based on the general philosophy of detection transformer, these methods implicitly learn the event locations and event semantics, which requires a large amount of training data and limits the model's performance in practice. In this paper, we propose a novel dense video captioning framework, named PR-DETR, which injects the explicit position and relation prior into the detection transformer to improve the localization accuracy and caption quality, simultaneously. On the one hand, we first generate a set of position-anchored queries to provide the scene-specific position and semantic information about potential events as position prior, which serves as the initial event search regions to eliminate the implausible event proposals. On the other hand, we further design an event relation encoder to explicitly calculate the relationship between event boundaries as relation prior to guide the event interaction to improve the semantic coherence of the captions. Extensive ablation studies are conducted to verify the effectiveness of the position and relation prior. Experimental results also show the competitive performance of our method on ActivityNet Captions and YouCook2 datasets.

cs.CV↗

arXiv2025

FlowRAM: Grounding Flow Matching Policy with Region-Aware Mamba Framework for Robotic Manipulation

Robotic manipulation in high-precision tasks is essential for numerous industrial and real-world applications where accuracy and speed are required. Yet current diffusion-based policy learning methods generally suffer from low computational efficiency due to the iterative denoising process during inference. Moreover, these methods do not fully explore the potential of generative models for enhancing information exploration in 3D environments. In response, we propose FlowRAM, a novel framework that leverages generative models to achieve region-aware perception, enabling efficient multimodal information processing. Specifically, we devise a Dynamic Radius Schedule, which allows adaptive perception, facilitating transitions from global scene comprehension to fine-grained geometric details. Furthermore, we integrate state space models to integrate multimodal information, while preserving linear computational complexity. In addition, we employ conditional flow matching to learn action poses by regressing deterministic vector fields, simplifying the learning process while maintaining performance. We verify the effectiveness of the FlowRAM in the RLBench, an established manipulation benchmark, and achieve state-of-the-art performance. The results demonstrate that FlowRAM achieves a remarkable improvement, particularly in high-precision tasks, where it outperforms previous methods by 12.0% in average success rate. Additionally, FlowRAM is able to generate physically plausible actions for a variety of real-world tasks in less than 4 time steps, significantly increasing inference speed.

cs.RO↗

arXiv2025

Time-Unified Diffusion Policy with Action Discrimination for Robotic Manipulation

In many complex scenarios, robotic manipulation relies on generative models to estimate the distribution of multiple successful actions. As the diffusion model has better training robustness than other generative models, it performs well in imitation learning through successful robot demonstrations. However, the diffusion-based policy methods typically require significant time to iteratively denoise robot actions, which hinders real-time responses in robotic manipulation. Moreover, existing diffusion policies model a time-varying action denoising process, whose temporal complexity increases the difficulty of model training and leads to suboptimal action accuracy. To generate robot actions efficiently and accurately, we present the Time-Unified Diffusion Policy (TUDP), which utilizes action recognition capabilities to build a time-unified denoising process. On the one hand, we build a time-unified velocity field in action space with additional action discrimination information. By unifying all timesteps of action denoising, our velocity field reduces the difficulty of policy learning and speeds up action generation. On the other hand, we propose an action-wise training method, which introduces an action discrimination branch to supply additional action discrimination information. Through action-wise training, the TUDP implicitly learns the ability to discern successful actions to better denoising accuracy. Our method achieves state-of-the-art performance on RLBench with the highest success rate of 82.6% on a multi-view setup and 83.8% on a single-view setup. In particular, when using fewer denoising iterations, TUDP achieves a more significant improvement in success rate. Additionally, TUDP can produce accurate actions for a wide range of real-world tasks.

cs.RO↗