Xiaopeng Fan

Wenxue Cui, Hualin Li, Yuhang Qin et al.

Recent deep unfolding networks (DUNs) have advanced Compressive Sensing (CS) by effectively integrating iterative optimization with deep learning architectures. However, most CS approaches predominantly confine their inference to a single solution space, neglecting the inherent ill-posedness of CS problems that intrinsically permits multiple plausible candidate hypotheses. In this paper, a novel Multi-Hypothesis Collaborative Deep Unfolding CS Network (MHC-DUN) is proposed, which explicitly models and leverages multiple hypotheses by jointly optimizing across diverse solution spaces. Specifically, following the Proximal Gradient Descent algorithm, MHC-DUN jointly performs gradient descent and proximal mapping within this multi-hypothesis paradigm. i) For gradient descent, a well-designed AlphaNet is introduced to dynamically predict spatially varying step sizes for all hypotheses, enabling collaborative gradient updates across multiple solutions. ii) For proximal operator, a sophisticated multi-hypothesis collaborative proximal mapping module is designed, which leverages both intra-hypothesis and inter-hypothesis correlation priors to jointly refine multiple solutions. To enable end-to-end training, a novel composite loss function is designed, which balances measurement fidelity, hypothesis diversity, and reconstruction accuracy, encouraging exploration of complementary solutions while maintaining reconstruction fidelity. Experimental results reveal that the proposed CS method outperforms existing CS networks.

Xu Liu, Jianing Li, Xiaopeng Fan et al.

Event cameras, offering high temporal resolutions and high dynamic ranges, have brought a new perspective to address common challenges (e.g., motion blur and low light) in monocular depth estimation. However, how to effectively exploit the sparse spatial information and rich temporal cues from asynchronous events remains a challenging endeavor. To this end, we propose a novel event-based monocular depth estimator with recurrent transformers, namely EReFormer, which is the first pure transformer with a recursive mechanism to process continuous event streams. Technically, for spatial modeling, a novel transformer-based encoder-decoder with a spatial transformer fusion module is presented, having better global context information modeling capabilities than CNN-based methods. For temporal modeling, we design a gate recurrent vision transformer unit that introduces a recursive mechanism into transformers, improving temporal modeling capabilities while alleviating the expensive GPU memory cost. The experimental results show that our EReFormer outperforms state-of-the-art methods by a margin on both synthetic and real-world datasets. We hope that our work will attract further research to develop stunning transformers in the event-based vision community. Our open-source code can be found in the supplemental material.

Wenbin Zou, Tianyi Li, Kejun Wu et al.

This paper reports on the NTIRE 2026 Challenge on Bitstream-Corrupted Video Restoration (BSCVR). The challenge aims to advance research on recovering visually coherent videos from corrupted bitstreams, whose decoding often produces severe spatial-temporal artifacts and content distortion. Built upon recent progress in bitstream-corrupted video recovery, the challenge provides a common benchmark for evaluating restoration methods under realistic corruption settings. We describe the dataset, evaluation protocol, and participating methods, and summarize the final results and main technical trends. The challenge highlights the difficulty of this emerging task and provides useful insights for future research on robust video restoration under practical bitstream corruption.

Chenhao Wang, Xiaopeng Hong, Zhiheng Ma et al.

Multi-modal crowd counting is a crucial task that uses multi-modal cues to estimate the number of people in crowded scenes. To overcome the gap between different modalities, we propose a modal emulation-based two-pass multi-modal crowd-counting framework that enables efficient modal emulation, alignment, and fusion. The framework consists of two key components: a \emph{multi-modal inference} pass and a \emph{cross-modal emulation} pass. The former utilizes a hybrid cross-modal attention module to extract global and local information and achieve efficient multi-modal fusion. The latter uses attention prompting to coordinate different modalities and enhance multi-modal alignment. We also introduce a modality alignment module that uses an efficient modal consistency loss to align the outputs of the two passes and bridge the semantic gap between modalities. Extensive experiments on both RGB-Thermal and RGB-Depth counting datasets demonstrate its superior performance compared to previous methods. Code available at https://github.com/Mr-Monday/Multi-modal-Crowd-Counting-via-Modal-Emulation.

Wenrui Li, Fucheng Cai, Yapeng Mi et al.

Text-driven 3D scene generation has seen significant advancements recently. However, most existing methods generate single-view images using generative models and then stitch them together in 3D space. This independent generation for each view often results in spatial inconsistency and implausibility in the 3D scenes. To address this challenge, we proposed a novel text-driven 3D-consistent scene generation model: SceneDreamer360. Our proposed method leverages a text-driven panoramic image generation model as a prior for 3D scene generation and employs 3D Gaussian Splatting (3DGS) to ensure consistency across multi-view panoramic images. Specifically, SceneDreamer360 enhances the fine-tuned Panfusion generator with a three-stage panoramic enhancement, enabling the generation of high-resolution, detail-rich panoramic images. During the 3D scene construction, a novel point cloud fusion initialization method is used, producing higher quality and spatially consistent point clouds. Our extensive experiments demonstrate that compared to other methods, SceneDreamer360 with its panoramic image generation and 3DGS can produce higher quality, spatially consistent, and visually appealing 3D scenes from any text prompt. Our codes are available at \url{https://github.com/liwrui/SceneDreamer360}.

Wenrui Li, Wei Han, Yandu Chen et al.

Due to the challenges in acquiring paired Text-3D data and the inherent irregularity of 3D data structures, combined representation learning of 3D point clouds and text remains unexplored. In this paper, we propose a novel Riemann-based Multi-scale Attention Reasoning Network (RMARN) for text-3D retrieval. Specifically, the extracted text and point cloud features are refined by their respective Adaptive Feature Refiner (AFR). Furthermore, we introduce the innovative Riemann Local Similarity (RLS) module and the Global Pooling Similarity (GPS) module. However, as 3D point cloud data and text data often possess complex geometric structures in high-dimensional space, the proposed RLS employs a novel Riemann Attention Mechanism to reflect the intrinsic geometric relationships of the data. Without explicitly defining the manifold, RMARN learns the manifold parameters to better represent the distances between text-point cloud samples. To address the challenges of lacking paired text-3D data, we have created the large-scale Text-3D Retrieval dataset T3DR-HIT, which comprises over 3,380 pairs of text and point cloud data. T3DR-HIT contains coarse-grained indoor 3D scenes and fine-grained Chinese artifact scenes, consisting of 1,380 and over 2,000 text-3D pairs, respectively. Experiments on our custom datasets demonstrate the superior performance of the proposed method. Our code and proposed datasets are available at \url{https://github.com/liwrui/RMARN}.

Xuehui Yu, Fucheng Cai, Meiyi Wang et al.

Inference-time guided sampling steers state-of-the-art diffusion and flow models without fine-tuning by interpreting the generation process as a controllable trajectory. This provides a simple and flexible way to inject external constraints (e.g., cost functions or pre-trained verifiers) for controlled generation. However, existing methods often fail when composing multiple constraints simultaneously, which leads to deviations from the true data manifold. In this work, we identify root causes of this off-manifold drift and find that the approximation error scales severely with gradient misalignment. Building on these findings, we propose Conflict-Aware Additive Guidance ($g^\text{car}$), a lightweight and learnable method, which actively rectifies off-manifold drift by dynamically detecting and resolving gradient conflicts. We validate $g^\text{car}$ across diverse domains, ranging from synthetic datasets and image editing to generative decision-making for planning and control. Our results demonstrate that $g^\text{car}$ effectively rectifies off-manifold drift, surpassing baselines in generation fidelity while using light compute. Code is available at https://github.com/yuxuehui/CAR-guidance.

Wenxue Cui, Xiaopeng Fan, Jian Zhang et al.

Inspired by certain optimization solvers, the deep unfolding network (DUN) has attracted much attention in recent years for image compressed sensing (CS). However, there still exist the following two issues: 1) In existing DUNs, most hyperparameters are usually content independent, which greatly limits their adaptability for different input contents. 2) In each iteration, a plain convolutional neural network is usually adopted, which weakens the perception of wider context prior and therefore depresses the expressive ability. In this paper, inspired by the traditional Proximal Gradient Descent (PGD) algorithm, a novel DUN for image compressed sensing (dubbed DUN-CSNet) is proposed to solve the above two issues. Specifically, for the first issue, a novel content adaptive gradient descent network is proposed, in which a well-designed step size generation sub-network is developed to dynamically allocate the corresponding step sizes for different textures of input image by generating a content-aware step size map, realizing a content-adaptive gradient updating. For the second issue, considering the fact that many similar patches exist in an image but have undergone a deformation, a novel deformation-invariant non-local proximal mapping network is developed, which can adaptively build the long-range dependencies between the nonlocal patches by deformation-invariant non-local modeling, leading to a wider perception on context priors. Extensive experiments manifest that the proposed DUN-CSNet outperforms existing state-of-the-art CS methods by large margins.

Jisheng Chu, Wenrui Li, Rui Zhao et al.

Generating immersive 3D scenes from texts is a core task in computer vision, crucial for applications in virtual reality and game development. Despite the promise of leveraging 2D diffusion priors, existing methods suffer from spatial blindness and rely on predefined trajectories that fail to exploit the inner relationships among salient objects. Consequently, these approaches are unable to comprehend the semantic layout, preventing them from exploring the scene adaptively to infer occluded content. Moreover, current inpainting models operate in 2D image space, struggling to plausibly fill holes caused by camera motion. To address these limitations, we propose RoamScene3D, a novel framework that bridges the gap between semantic guidance and spatial generation. Our method reasons about the semantic relations among objects and produces consistent and photorealistic scenes. Specifically, we employ a vision-language model (VLM) to construct a scene graph that encodes object relations, guiding the camera to perceive salient object boundaries and plan an adaptive roaming trajectory. Furthermore, to mitigate the limitations of static 2D priors, we introduce a Motion-Injected Inpainting model that is fine-tuned on a synthetic panoramic dataset integrating authentic camera trajectories, making it adaptive to camera motion. Extensive experiments demonstrate that with semantic reasoning and geometric constraints, our method significantly outperforms state-of-the-art approaches in producing consistent and photorealistic scenes. Our code is available at https://github.com/JS-CHU/RoamScene3D.

Wenrui Li, Hongtao Chen, Yao Xiao et al.

All-in-one image restoration aims to recover clean images from diverse unknown degradations using a single model. But extending this task to videos faces unique challenges. Existing approaches primarily focus on frame-wise degradation variation, overlooking the temporal continuity that naturally exists in real-world degradation processes. In practice, degradation types and intensities evolve smoothly over time, and multiple degradations may coexist or transition gradually. In this paper, we introduce the Smoothly Evolving Unknown Degradations (SEUD) scenario, where both the active degradation set and degradation intensity change continuously over time. To support this scenario, we design a flexible synthesis pipeline that generates temporally coherent videos with single, compound, and evolving degradations. To address the challenges in the SEUD scenario, we propose an all-in-One Recurrent Conditional and Adaptive prompting Network (ORCANet). First, a Coarse Intensity Estimation Dehazing (CIED) module estimates haze intensity using physical priors and provides coarse dehazed features as initialization. Second, a Flow Prompt Generation (FPG) module extracts degradation features. FPG generates both static prompts that capture segment-level degradation types and dynamic prompts that adapt to frame-level intensity variations. Furthermore, a label-aware supervision mechanism improves the discriminability of static prompt representations under different degradations. Extensive experiments show that ORCANet achieves superior restoration quality, temporal consistency, and robustness over image and video-based baselines. Code is available at https://github.com/Friskknight/ORCANet-SEUD.

Wenrui Li, Zhengyu Ma, Jinqiao Shi et al.

Image-text retrieval which associates different modalities has drawn broad attention due to its excellent research value and broad real-world application. However, most of the existing methods haven't taken the high-level semantic relationships ("style embedding") and common knowledge from multi-modalities into full consideration. To this end, we introduce a novel style transformer network with common knowledge optimization (CKSTN) for image-text retrieval. The main module is the common knowledge adaptor (CKA) with both the style embedding extractor (SEE) and the common knowledge optimization (CKO) modules. Specifically, the SEE uses the sequential update strategy to effectively connect the features of different stages in SEE. The CKO module is introduced to dynamically capture the latent concepts of common knowledge from different modalities. Besides, to get generalized temporal common knowledge, we propose a sequential update strategy to effectively integrate the features of different layers in SEE with previous common feature units. CKSTN demonstrates the superiorities of the state-of-the-art methods in image-text retrieval on MSCOCO and Flickr30K datasets. Moreover, CKSTN is constructed based on the lightweight transformer which is more convenient and practical for the application of real scenes, due to the better performance and lower parameters.

Wenrui Li, Yidan Lu, Yeyu Chai et al.

With the daily influx of 3D data on the internet, text-3D retrieval has gained increasing attention. However, current methods face two major challenges: Hierarchy Representation Collapse (HRC) and Redundancy-Induced Saliency Dilution (RISD). HRC compresses abstract-to-specific and whole-to-part hierarchies in Euclidean embeddings, while RISD averages noisy fragments, obscuring critical semantic cues and diminishing the model's ability to distinguish hard negatives. To address these challenges, we introduce the Hyperbolic Hierarchical Alignment Reasoning Network (H$^{2}$ARN) for text-3D retrieval. H$^{2}$ARN embeds both text and 3D data in a Lorentz-model hyperbolic space, where exponential volume growth inherently preserves hierarchical distances. A hierarchical ordering loss constructs a shrinking entailment cone around each text vector, ensuring that the matched 3D instance falls within the cone, while an instance-level contrastive loss jointly enforces separation from non-matching samples. To tackle RISD, we propose a contribution-aware hyperbolic aggregation module that leverages Lorentzian distance to assess the relevance of each local feature and applies contribution-weighted aggregation guided by hyperbolic geometry, enhancing discriminative regions while suppressing redundancy without additional supervision. We also release the expanded T3DR-HIT v2 benchmark, which contains 8,935 text-to-3D pairs, 2.6 times the original size, covering both fine-grained cultural artefacts and complex indoor scenes. Our codes are available at https://github.com/liwrui/H2ARN.

Zhe Yang, Wenrui Li, Hongtao Chen et al.

Multimodal learning aims to improve performance by leveraging data from multiple sources. During joint multimodal training, due to modality bias, the advantaged modality often dominates backpropagation, leading to imbalanced optimization. Existing methods still face two problems: First, the long-term dominance of the dominant modality weakens representation-output coupling in the late stages of training, resulting in the accumulation of redundant information. Second, previous methods often directly and uniformly adjust the gradients of the advantaged modality, ignoring the semantics and directionality between modalities. To address these limitations, we propose Adaptive Redundancy Regulation for Balanced Multimodal Information Refinement (RedReg), which is inspired by information bottleneck principle. Specifically, we construct a redundancy phase monitor that uses a joint criterion of effective gain growth rate and redundancy to trigger intervention only when redundancy is high. Furthermore, we design a co-information gating mechanism to estimate the contribution of the current dominant modality based on cross-modal semantics. When the task primarily relies on a single modality, the suppression term is automatically disabled to preserve modality-specific information. Finally, we project the gradient of the dominant modality onto the orthogonal complement of the joint multimodal gradient subspace and suppress the gradient according to redundancy. Experiments show that our method demonstrates superiority among current major methods in most scenarios. Ablation experiments verify the effectiveness of our method. The code is available at https://github.com/xia-zhe/RedReg.git

Wenrui Li, Wei Han, Hengyu Man et al.

With the rapid growth of video content on social media, video summarization has become a crucial task in multimedia processing. However, existing methods face challenges in capturing global dependencies in video content and accommodating multimodal user customization. Moreover, temporal proximity between video frames does not always correspond to semantic proximity. To tackle these challenges, we propose a novel Language-guided Graph Representation Learning Network (LGRLN) for video summarization. Specifically, we introduce a video graph generator that converts video frames into a structured graph to preserve temporal order and contextual dependencies. By constructing forward, backward and undirected graphs, the video graph generator effectively preserves the sequentiality and contextual relationships of video content. We designed an intra-graph relational reasoning module with a dual-threshold graph convolution mechanism, which distinguishes semantically relevant frames from irrelevant ones between nodes. Additionally, our proposed language-guided cross-modal embedding module generates video summaries with specific textual descriptions. We model the summary generation output as a mixture of Bernoulli distribution and solve it with the EM algorithm. Experimental results show that our method outperforms existing approaches across multiple benchmarks. Moreover, we proposed LGRLN reduces inference time and model parameters by 87.8% and 91.7%, respectively. Our codes and pre-trained models are available at https://github.com/liwrui/LGRLN.

Xianqi Zhang, Xingtao Wang, Xu Liu et al.

Reinforcement Learning (RL) and Imitation Learning (IL) have made great progress in robotic decision-making in recent years. However, these methods show obvious deterioration for new tasks that need to be completed through new combinations of actions. RL methods suffer from reward functions and distribution shifts, while IL methods are limited by expert demonstrations which do not cover new tasks. In contrast, humans can easily complete these tasks with the fragmented knowledge learned from task-agnostic experience. Inspired by this observation, this paper proposes a task-agnostic learning method (TAL for short) that can learn fragmented knowledge only from task-agnostic data to accomplish new tasks. TAL consists of four stages. First, the task-agnostic exploration is performed to collect data from interactions with the environment. The collected data is organized via a knowledge graph. Second, an action feature extractor is proposed and trained using the collected knowledge graph data for task-agnostic fragmented knowledge learning. Third, a candidate action generator is designed, which applies the action feature extractor on a new task to generate multiple candidate action sets. Finally, an action proposal network is designed to produce the probabilities for actions in a new task according to the environmental information. The probabilities are then used to generate order information for selecting actions to be executed from multiple candidate action sets to form the plan. Experiments on a virtual indoor scene show that the proposed method outperforms the state-of-the-art offline RL methods and IL methods by more than 20%.

Zhaolin Wan, Yining Diao, Jingqi Xu et al.

With the rapid advancement of 3D visualization, 3D Gaussian Splatting (3DGS) has emerged as a leading technique for real-time, high-fidelity rendering. While prior research has emphasized algorithmic performance and visual fidelity, the perceptual quality of 3DGS-rendered content, especially under varying reconstruction conditions, remains largely underexplored. In practice, factors such as viewpoint sparsity, limited training iterations, point downsampling, noise, and color distortions can significantly degrade visual quality, yet their perceptual impact has not been systematically studied. To bridge this gap, we present 3DGS-QA, the first subjective quality assessment dataset for 3DGS. It comprises 225 degraded reconstructions across 15 object types, enabling a controlled investigation of common distortion factors. Based on this dataset, we introduce a no-reference quality prediction model that directly operates on native 3D Gaussian primitives, without requiring rendered images or ground-truth references. Our model extracts spatial and photometric cues from the Gaussian representation to estimate perceived quality in a structure-aware manner. We further benchmark existing quality assessment methods, spanning both traditional and learning-based approaches. Experimental results show that our method consistently achieves superior performance, highlighting its robustness and effectiveness for 3DGS content evaluation. The dataset and code are made publicly available at https://github.com/diaoyn/3DGSQA to facilitate future research in 3DGS quality assessment.

Wenrui Li, Penghong Wang, Ruiqin Xiong et al.

The spiking neural networks (SNNs) that efficiently encode temporal sequences have shown great potential in extracting audio-visual joint feature representations. However, coupling SNNs (binary spike sequences) with transformers (float-point sequences) to jointly explore the temporal-semantic information still facing challenges. In this paper, we introduce a novel Spiking Tucker Fusion Transformer (STFT) for audio-visual zero-shot learning (ZSL). The STFT leverage the temporal and semantic information from different time steps to generate robust representations. The time-step factor (TSF) is introduced to dynamically synthesis the subsequent inference information. To guide the formation of input membrane potentials and reduce the spike noise, we propose a global-local pooling (GLP) which combines the max and average pooling operations. Furthermore, the thresholds of the spiking neurons are dynamically adjusted based on semantic and temporal cues. Integrating the temporal and semantic information extracted by SNNs and Transformers are difficult due to the increased number of parameters in a straightforward bilinear model. To address this, we introduce a temporal-semantic Tucker fusion module, which achieves multi-scale fusion of SNN and Transformer outputs while maintaining full second-order interactions. Our experimental results demonstrate the effectiveness of the proposed approach in achieving state-of-the-art performance in three benchmark datasets. The harmonic mean (HM) improvement of VGGSound, UCF101 and ActivityNet are around 15.4\%, 3.9\%, and 14.9\%, respectively.

Chenlin Zhou, Han Zhang, Zhaokun Zhou et al.

Spiking Transformers, which integrate Spiking Neural Networks (SNNs) with Transformer architectures, have attracted significant attention due to their potential for energy efficiency and high performance. However, existing models in this domain still suffer from suboptimal performance. We introduce several innovations to improve the performance: i) We propose a novel spike-form Q-K attention mechanism, tailored for SNNs, which efficiently models the importance of token or channel dimensions through binary vectors with linear complexity. ii) We incorporate the hierarchical structure, which significantly benefits the performance of both the brain and artificial neural networks, into spiking transformers to obtain multi-scale spiking representation. iii) We design a versatile and powerful patch embedding module with a deformed shortcut specifically for spiking transformers. Together, we develop QKFormer, a hierarchical spiking transformer based on Q-K attention with direct training. QKFormer shows significantly superior performance over existing state-of-the-art SNN models on various mainstream datasets. Notably, with comparable size to Spikformer (66.34 M, 74.81%), QKFormer (64.96 M) achieves a groundbreaking top-1 accuracy of 85.65% on ImageNet-1k, substantially outperforming Spikformer by 10.84%. To our best knowledge, this is the first time that directly training SNNs have exceeded 85% accuracy on ImageNet-1K. The code and models are publicly available at https://github.com/zhouchenlin2096/QKFormer

Boyi Pang, Savva Ignatyev, Vladimir Ippolitov et al.

We suggest a new multi-modal algorithm for joint inference of paired structurally aligned samples with Rectified Flow models. While some existing methods propose a codependent generation process, they do not view the problem of joint generation from a structural alignment perspective. Recent work uses Score Distillation Sampling to generate aligned 3D models, but SDS is known to be time-consuming, prone to mode collapse, and often provides cartoonish results. By contrast, our suggested approach relies on the joint transport of a segment in the sample space, yielding faster computation at inference time. Our approach can be built on top of an arbitrary Rectified Flow model operating on the structured latent space. We show the applicability of our method to the domains of image, video, and 3D shape generation using state-of-the-art baselines and evaluate it against both editing-based and joint inference-based competing approaches. We demonstrate a high degree of structural alignment for the sample pairs obtained with our method and a high visual quality of the samples. Our method improves the state-of-the-art for image and video generation pipelines. For 3D generation, it is able to show comparable quality while working orders of magnitude faster.

Chunwei Tian, Mingjian Song, Xiaopeng Fan et al.

Deep convolutional neural networks can extract more accurate structural information via deep architectures to obtain good performance in image super-resolution. However, it is not easy to find effect of important layers in a single network architecture to decrease performance of super-resolution. In this paper, we design a tree-guided CNN for image super-resolution (TSRNet). It uses a tree architecture to guide a deep network to enhance effect of key nodes to amplify the relation of hierarchical information for improving the ability of recovering images. To prevent insufficiency of the obtained structural information, cosine transform techniques in the TSRNet are used to extract cross-domain information to improve the performance of image super-resolution. Adaptive Nesterov momentum optimizer (Adan) is applied to optimize parameters to boost effectiveness of training a super-resolution model. Extended experiments can verify superiority of the proposed TSRNet for restoring high-quality images. Its code can be obtained at https://github.com/hellloxiaotian/TSRNet.

Dianze Li, Jianing Li, Xu Liu et al.

Combining the complementary benefits of frames and events has been widely used for object detection in challenging scenarios. However, most object detection methods use two independent Artificial Neural Network (ANN) branches, limiting cross-modality information interaction across the two visual streams and encountering challenges in extracting temporal cues from event streams with low power consumption. To address these challenges, we propose HDI-Former, a Hybrid Dynamic Interaction ANN-SNN Transformer, marking the first trial to design a directly trained hybrid ANN-SNN architecture for high-accuracy and energy-efficient object detection using frames and events. Technically, we first present a novel semantic-enhanced self-attention mechanism that strengthens the correlation between image encoding tokens within the ANN Transformer branch for better performance. Then, we design a Spiking Swin Transformer branch to model temporal cues from event streams with low power consumption. Finally, we propose a bio-inspired dynamic interaction mechanism between ANN and SNN sub-networks for cross-modality information interaction. The results demonstrate that our HDI-Former outperforms eleven state-of-the-art methods and our four baselines by a large margin. Our SNN branch also shows comparable performance to the ANN with the same architecture while consuming 10.57$\times$ less energy on the DSEC-Detection dataset. Our open-source code is available in the supplementary material.

Xiangyang Hu, Xiangyu Shen, Yifei Sun et al.

Alzheimer's disease (AD) is a common neurodegenerative disease among the elderly. Early prediction and timely intervention of its prodromal stage, mild cognitive impairment (MCI), can decrease the risk of advancing to AD. Combining information from various modalities can significantly improve predictive accuracy. However, challenges such as missing data and heterogeneity across modalities complicate multimodal learning methods as adding more modalities can worsen these issues. Current multimodal fusion techniques often fail to adapt to the complexity of medical data, hindering the ability to identify relationships between modalities. To address these challenges, we propose an innovative multimodal approach for predicting MCI conversion, focusing specifically on the issues of missing positron emission tomography (PET) data and integrating diverse medical information. The proposed incomplete triple-modal MCI conversion prediction network is tailored for this purpose. Through the missing modal generation module, we synthesize the missing PET data from the magnetic resonance imaging and extract features using specifically designed encoders. We also develop a channel aggregation module and a triple-modal co-attention fusion module to reduce feature redundancy and achieve effective multimodal data fusion. Furthermore, we design a loss function to handle missing modality issues and align cross-modal features. These components collectively harness multimodal data to boost network performance. Experimental results on the ADNI1 and ADNI2 datasets show that our method significantly surpasses existing unimodal and other multimodal models. Our code is available at https://github.com/justinhxy/ITFC.

Xuan Deng, Xiandong Meng, Hengyu Man et al.

Although 3D Gaussian Splatting (3DGS) enables high-fidelity real-time rendering, its prohibitive storage overhead severely hinders practical deployment. Recent anchor-based 3DGS compression schemes reduce gaussian redundancy through some advanced context models. However, they overlook explicit geometric dependencies, leading to structural degradation and suboptimal ratedistortion performance. In this paper, we propose a Local Geometry-aware Hierarchical Context Compression framework for 3DGS(LG-HCC) that incorporates inter-anchor geometric correlations into anchor pruning and entropy coding for compact representation. Specifically, we introduce an Neighborhood-Aware Anchor Pruning (NAAP) strategy, which evaluates anchor importance via weighted neighborhood feature aggregation and then merges low-contribution anchors into salient neighbors, yielding a compact yet geometry-consistent anchor set. Moreover, we further develop a hierarchical entropy coding scheme, in which coarse-to-fine priors are exploited through a lightweight Geometry-Guided Convolution(GG-Conv) operator to enable spatially adaptive context modeling and rate-distortion optimization. Extensive experiments show that LG-HCC effectively alleviates structural preservation issues,achieving superior geometric integrity and rendering fidelity while reducing storage by up to 30.85x compared to the Scaffold-GS baseline on the Mip-NeRF360 dataset

Kanglin Ning, Yiran Zhao, Wenrui Li et al.

Three-dimensional object detection in panoramic imagery is crucial for comprehensive scene understanding, yet accurately mapping 2D features to 3D remains a significant challenge. Prevailing methods often project 2D features onto discrete 3D grids, which break geometric continuity and limit representation efficiency. To overcome this limitation, this paper proposes PanoGSDet, a monocular panoramic 3D detection framework built upon continuous semantic 3D Gaussian representations. The proposed framework comprises a panoramic depth estimation component and a semantic Gaussian component. The panoramic depth estimation component extracts the equirectangular semantic and depth features from the monocular panorama input. The semantic Gaussian component includes a semantic Gaussian lifting module that projects spherical features into 3D semantic Gaussians, a semantic Gaussian optimization module that refines these semantic Gaussians, and a Gaussian guided prediction head that generates 3D bounding boxes from optimized Gaussian representations. Extensive experiments on the Structured3D dataset demonstrate that our method significantly outperforms existing methods.

Hantang Li, Qiang Zhu, Xiandong Meng et al.

Dropout-based sparse-view 3D Gaussian Splatting (3DGS) methods alleviate overfitting by randomly suppressing Gaussian primitives during training. Existing methods mainly focus on designing increasingly sophisticated dropout strategies, while they overlook the resulting inconsistencies among different dropped Gaussian subsets. This oversight often leads to unstable reconstruction and suboptimal Gaussian representation learning.In this paper, we revisit dropout-based sparse-view 3DGS from a consistency regularization perspective and propose PairDropGS, a Paired Dropout-induced Consistency Regularization framework for sparse-view Gaussian splatting. Specifically, PairDropGS first constructs a pair of the dropped Gaussian subsets from a shared Gaussian field and designs a low-frequency consistency regularization to constrain their low-frequency rendered structures. This design encourages the shared Gaussian field to preserve stable scene layout and coarse geometry under different random dropouts, while avoiding excessive constraints on ambiguous high-frequency details. Moreover, we introduce a progressive consistency scheduling strategy to gradually strengthen the consistency regularization during training for stability and robustness of reconstruction. Extensive experiments on widely-used sparse-view benchmarks demonstrate that PairDropGS achieves superior training stability, significantly outperforms existing dropout-based 3DGS methods in reconstruction quality, while exhibiting the simplicity and plug-and-play nature for improving dropout-based optimization.

Kanglin Ning, Wenrui Li, Houde Quan et al.

Cross-modal knowledge distillation has emerged as an effective strategy for integrating point cloud and image features in 3D perception tasks. However, the modality heterogeneity, spatial misalignment, and the representation crisis of multiple modalities often limit the efficient of these cross-modal distillation methods. To address these limitations in existing approaches, we propose a hyperbolic constrained cross-modal distillation method for multimodal 3D object detection (HGC-Det). The proposed HGC-Det framework includes an image branch and a point cloud branch to extract semantic features from two different modalities. The point cloud branch comprises three core components: a 2D semantic-guided voxel optimization component (SGVO), a hyperbolic geometry constrained cross-modal feature transfer component (HFT), and a feature aggregation-based geometry optimization component (FAGO). Specifically, the SGVO component adaptively refines the spatial representation of the 3D branch by leveraging semantic cues from the image branch, thereby mitigating the issue of inadequate representation fusion. The HFT component exploits the intrinsic geometric properties of hyperbolic space to alleviate semantic loss during the fusion of high-dimensional image features and low-dimensional point cloud features. Finally, the FAGO compensates for potential spatial feature degradation introduced by the 2D semantic-guided voxel optimization component. Extensive experiments on indoor datasets (SUN RGB-D, ARKitScenes) and outdoor datasets (KITTI, nuScenes) demonstrate that our method achieves a better trade-off between detection accuracy and computational cost.

Kanglin Ning, Ruzhao Chen, Penghong Wang et al.

Predicting spherical pixel depth from monocular $360^{\circ}$ indoor panoramas is critical for many vision applications. However, existing methods focus on pixel-level accuracy, causing oversmoothed room corners and noise sensitivity. In this paper, we propose a depth estimation framework based on room geometry constraints, which extracts room geometry information through layout prediction and integrates those information into the depth estimation process through background segmentation mechanism. At the model level, our framework comprises a shared feature encoder followed by task-specific decoders for layout estimation, depth estimation, and background segmentation. The shared encoder extracts multi-scale features, which are subsequently processed by individual decoders to generate initial predictions: a depth map, a room layout map, and a background segmentation map. Furthermore, our framework incorporates two strategies: a room geometry-based background depth resolving strategy and a background-segmentation-guided fusion mechanism. The proposed room-geometry-based background depth resolving strategy leverages the room layout and the depth decoder's output to generate the corresponding background depth map. Then, a background-segmentation-guided fusion strategy derives fusion weights for the background and coarse depth maps from the segmentation decoder's predictions. Extensive experimental results on the Stanford2D3D, Matterport3D and Structured3D datasets show that our proposed methods can achieve significantly superior performance than current open-source methods. Our code is available at https://github.com/emiyaning/RGCNet.

Wenrui Li, Wei Han, Liang-Jian Deng et al.

With the rise of short video content, efficient video summarization techniques for extracting key information have become crucial. However, existing methods struggle to capture the global temporal dependencies and maintain the semantic coherence of video content. Additionally, these methods are also influenced by noise during multi-channel feature fusion. We propose a Spiking Variational Graph (SpiVG) Network, which enhances information density and reduces computational complexity. First, we design a keyframe extractor based on Spiking Neural Networks (SNN), leveraging the event-driven computation mechanism of SNNs to learn keyframe features autonomously. To enable fine-grained and adaptable reasoning across video frames, we introduce a Dynamic Aggregation Graph Reasoner, which decouples contextual object consistency from semantic perspective coherence. We present a Variational Inference Reconstruction Module to address uncertainty and noise arising during multi-channel feature fusion. In this module, we employ Evidence Lower Bound Optimization (ELBO) to capture the latent structure of multi-channel feature distributions, using posterior distribution regularization to reduce overfitting. Experimental results show that SpiVG surpasses existing methods across multiple datasets such as SumMe, TVSum, VideoXum, and QFVS. Our codes and pre-trained models are available at https://github.com/liwrui/SpiVG.

Hongliang Wei, Xianqi Zhang, Xingtao Wang et al.

Despite significant progress, existing research on Multimodal Large Language Models (MLLMs) mainly focuses on general visual understanding, overlooking the ability to integrate textual context associated with objects for a more context-aware multimodal understanding -- an ability we refer to as Region-level Context-aware Multimodal Understanding (RCMU). To address this limitation, we first formulate the RCMU task, which requires models to respond to user instructions by integrating both image content and textual information of regions or objects. To equip MLLMs with RCMU capabilities, we propose Region-level Context-aware Visual Instruction Tuning (RCVIT), which incorporates object information into the model input and enables the model to utilize bounding box coordinates to effectively associate objects' visual content with their textual information. To address the lack of datasets, we introduce the RCMU dataset, a large-scale visual instruction tuning dataset that covers multiple RCMU tasks. We also propose RC\&P-Bench, a comprehensive benchmark that can evaluate the performance of MLLMs in RCMU and multimodal personalized understanding tasks. Additionally, we propose a reference-free evaluation metric to perform a comprehensive and fine-grained evaluation of the region-level context-aware image descriptions. By performing RCVIT on Qwen2-VL models with the RCMU dataset, we developed RC-Qwen2-VL models. Experimental results indicate that RC-Qwen2-VL models not only achieve outstanding performance on multiple RCMU tasks but also demonstrate successful applications in multimodal RAG and personalized conversation. Our data, model and benchmark are available at https://github.com/hongliang-wei/RC-MLLM

Jisheng Chu, Wenrui Li, Xingtao Wang et al.

The common occurrence of occlusion-induced incompleteness in point clouds has made point cloud completion (PCC) a highly-concerned task in the field of geometric processing. Existing PCC methods typically produce complete point clouds from partial point clouds in a coarse-to-fine paradigm, with the coarse stage generating entire shapes and the fine stage improving texture details. Though diffusion models have demonstrated effectiveness in the coarse stage, the fine stage still faces challenges in producing high-fidelity results due to the ill-posed nature of PCC. The intrinsic contextual information for texture details in partial point clouds is the key to solving the challenge. In this paper, we propose a high-fidelity PCC method that digs into both short and long-range contextual information from the partial point cloud in the fine stage. Specifically, after generating the coarse point cloud via a diffusion-based coarse generator, a mixed sampling module introduces short-range contextual information from partial point clouds into the fine stage. A surface freezing modules safeguards points from noise-free partial point clouds against disruption. As for the long-range contextual information, we design a similarity modeling module to derive similarity with rigid transformation invariance between points, conducting effective matching of geometric manifold features globally. In this way, the high-quality components present in the partial point cloud serve as valuable references for refining the coarse point cloud with high fidelity. Extensive experiments have demonstrated the superiority of the proposed method over SOTA competitors. Our code is available at https://github.com/JS-CHU/ContextualCompletion.

Chong Mou, Jian Zhang, Xiaopeng Fan et al.

Local and non-local attention-based methods have been well studied in various image restoration tasks while leading to promising performance. However, most of the existing methods solely focus on one type of attention mechanism (local or non-local). Furthermore, by exploiting the self-similarity of natural images, existing pixel-wise non-local attention operations tend to give rise to deviations in the process of characterizing long-range dependence due to image degeneration. To overcome these problems, in this paper we propose a novel collaborative attention network (COLA-Net) for image restoration, as the first attempt to combine local and non-local attention mechanisms to restore image content in the areas with complex textures and with highly repetitive details respectively. In addition, an effective and robust patch-wise non-local attention model is developed to capture long-range feature correspondences through 3D patches. Extensive experiments on synthetic image denoising, real image denoising and compression artifact reduction tasks demonstrate that our proposed COLA-Net is able to achieve state-of-the-art performance in both peak signal-to-noise ratio and visual perception, while maintaining an attractive computational complexity. The source code is available on https://github.com/MC-E/COLA-Net.

Wenrui Li, Xiaopeng Hong, Ruiqin Xiong et al.

Temporal video grounding (TVG) is a critical task in video content understanding, requiring precise alignment between video content and natural language instructions. Despite significant advancements, existing methods face challenges in managing confidence bias towards salient objects and capturing long-term dependencies in video sequences. To address these issues, we introduce SpikeMba: a multi-modal spiking saliency mamba for temporal video grounding. Our approach integrates Spiking Neural Networks (SNNs) with state space models (SSMs) to leverage their unique advantages in handling different aspects of the task. Specifically, we use SNNs to develop a spiking saliency detector that generates the proposal set. The detector emits spike signals when the input signal exceeds a predefined threshold, resulting in a dynamic and binary saliency proposal set. To enhance the model's capability to retain and infer contextual information, we introduce relevant slots which learnable tensors that encode prior knowledge. These slots work with the contextual moment reasoner to maintain a balance between preserving contextual information and exploring semantic relevance dynamically. The SSMs facilitate selective information propagation, addressing the challenge of long-term dependency in video content. By combining SNNs for proposal generation and SSMs for effective contextual reasoning, SpikeMba addresses confidence bias and long-term dependencies, thereby significantly enhancing fine-grained multimodal relationship capture. Our experiments demonstrate the effectiveness of SpikeMba, which consistently outperforms state-of-the-art methods across mainstream benchmarks.

Hantang Li, Qiang Zhu, Xiandong Meng et al.

Sparse-view reconstruction with 3D Gaussian Splatting (3DGS) is fundamentally ill-posed due to insufficient geometric supervision, often leading to severe overfitting and the emergence of structural distortions and translucent haze-like artifacts. While existing approaches attempt to alleviate this issue via dropout-based regularization, they are largely heuristic and lack a unified understanding of artifact formation. In this paper, we revisit sparse-view 3DGS reconstruction from a new perspective and identify the core challenge as the unobservability of Gaussian primitive reliability. Unreliable Gaussians are insufficiently constrained during optimization and accumulate as haze-like degradations in rendered images. Motivated by this observation, we propose a unified Dual-domain Observation and Calibration (DOC-GS) framework that models and corrects Gaussian reliability through the synergy of optimization-domain inductive bias and observation-domain evidence. Specifically, in the optimization domain, we characterize Gaussian reliability by the degree to which each primitive is constrained during training, and instantiate this signal via a Continuous Depth-Guided Dropout (CDGD) strategy, where the dropout probability serves as an explicit proxy for primitive reliability. This imposes a smooth depth-aware inductive bias to suppress weakly constrained Gaussians and improve optimization stability. In the observation domain, we establish a connection between floater artifacts and atmospheric scattering, and leverage the Dark Channel Prior (DCP) as a structural consistency cue to identify and accumulate anomalous regions. Based on cross-view aggregated evidence, we further design a reliability-driven geometric pruning strategy to remove low-confidence Gaussians.

Shuhong Liu, Chenyu Bao, Ziteng Cui et al.

This paper presents a comprehensive review of the NTIRE 2026 3D Restoration and Reconstruction (3DRR) Challenge, detailing the proposed methods and results. The challenge seeks to identify robust reconstruction pipelines that are robust under real-world adverse conditions, specifically extreme low-light and smoke-degraded environments, as captured by our RealX3D benchmark. A total of 279 participants registered for the competition, of whom 33 teams submitted valid results. We thoroughly evaluate the submitted approaches against state-of-the-art baselines, revealing significant progress in 3D reconstruction under adverse conditions. Our analysis highlights shared design principles among top-performing methods and provides insights into effective strategies for handling 3D scene degradation.

Wenxue Cui, Xingtao Wang, Xiaopeng Fan et al.

Existing image compressed sensing (CS) coding frameworks usually solve an inverse problem based on measurement coding and optimization-based image reconstruction, which still exist the following two challenges: 1) The widely used random sampling matrix, such as the Gaussian Random Matrix (GRM), usually leads to low measurement coding efficiency. 2) The optimization-based reconstruction methods generally maintain a much higher computational complexity. In this paper, we propose a new CNN based image CS coding framework using local structural sampling (dubbed CSCNet) that includes three functional modules: local structural sampling, measurement coding and Laplacian pyramid reconstruction. In the proposed framework, instead of GRM, a new local structural sampling matrix is first developed, which is able to enhance the correlation between the measurements through a local perceptual sampling strategy. Besides, the designed local structural sampling matrix can be jointly optimized with the other functional modules during training process. After sampling, the measurements with high correlations are produced, which are then coded into final bitstreams by the third-party image codec. At last, a Laplacian pyramid reconstruction network is proposed to efficiently recover the target image from the measurement domain to the image domain. Extensive experimental results demonstrate that the proposed scheme outperforms the existing state-of-the-art CS coding methods, while maintaining fast computational speed.

Wenrui Li, Zhe Yang, Wei Han et al.

Reconstructing desired objects and scenes has long been a primary goal in 3D computer vision. Single-view point cloud reconstruction has become a popular technique due to its low cost and accurate results. However, single-view reconstruction methods often rely on expensive CAD models and complex geometric priors. Effectively utilizing prior knowledge about the data remains a challenge. In this paper, we introduce hyperbolic space to 3D point cloud reconstruction, enabling the model to represent and understand complex hierarchical structures in point clouds with low distortion. We build upon previous methods by proposing a hyperbolic Chamfer distance and a regularized triplet loss to enhance the relationship between partial and complete point clouds. Additionally, we design adaptive boundary conditions to improve the model's understanding and reconstruction of 3D structures. Our model outperforms most existing models, and ablation studies demonstrate the significance of our model and its components. Experimental results show that our method significantly improves feature extraction capabilities. Our model achieves outstanding performance in 3D reconstruction tasks.

Xiang Tang, Ruotong Li, Xiaopeng Fan

In recent years, the demand for 3D content has grown exponentially with intelligent upgrading of interactive media, extended reality (XR), and Metaverse industries. In order to overcome the limitation of traditional manual modeling approaches, such as labor-intensive workflows and prolonged production cycles, revolutionary advances have been achieved through the convergence of novel 3D representation paradigms and artificial intelligence generative technologies. In this survey, we conduct a systematically review of the cutting-edge achievements in static 3D object and scene generation, as well as establish a comprehensive technical framework through systematic categorization. Specifically, we initiate our analysis with mainstream 3D object representations, followed by in-depth exploration of two principal technical pathways in object generation: data-driven supervised learning methods and deep generative model-based approaches. Regarding scene generation, we focus on three dominant paradigms: layout-guided compositional synthesis, 2D prior-based scene generation, and rule-driven modeling. Finally, we critically examine persistent challenges in 3D generation and propose potential research directions for future investigation. This survey aims to provide readers with a structured understanding of state-of-the-art 3D generation technologies while inspiring researchers to undertake more exploration in this domain.

Xiang Tang, Ruotong Li, Xiaopeng Fan

In recent years, 3D generation has made great strides in both academia and industry. However, generating 3D scenes from a single RGB image remains a significant challenge, as current approaches often struggle to ensure both object generation quality and scene coherence in multi-object scenarios. To overcome these limitations, we propose a novel three-stage framework for 3D scene generation with explicit geometric representations and high-quality textural details via single image-guided model generation and spatial layout optimization. Our method begins with an image instance segmentation and inpainting phase, which recovers missing details of occluded objects in the input images, thereby achieving complete generation of foreground 3D assets. Subsequently, our approach captures the spatial geometry of reference image by constructing pseudo-stereo viewpoint for camera parameter estimation and scene depth inference, while employing a model selection strategy to ensure optimal alignment between the 3D assets generated in the previous step and the input. Finally, through model parameterization and minimization of the Chamfer distance between point clouds in 3D and 2D space, our approach optimizes layout parameters to produce an explicit 3D scene representation that maintains precise alignment with input guidance image. Extensive experiments on multi-object scene image sets have demonstrated that our approach not only outperforms state-of-the-art methods in terms of geometric accuracy and texture fidelity of individual generated 3D models, but also has significant advantages in scene layout synthesis.

Zhitao Wang, Hengyu Man, Wenrui Li et al.

Recent advances in video generation techniques have given rise to an emerging paradigm of generative video coding for Ultra-Low Bitrate (ULB) scenarios by leveraging powerful generative priors. However, most existing methods are limited by domain specificity (e.g., facial or human videos) or excessive dependence on high-level text guidance, which tend to inadequately capture fine-grained motion details, leading to unrealistic or incoherent reconstructions. To address these challenges, we propose Trajectory-Guided Generative Video Coding (dubbed T-GVC), a novel framework that bridges low-level motion tracking with high-level semantic understanding. T-GVC features a semantic-aware sparse motion sampling pipeline that extracts pixel-wise motion as sparse trajectory points based on their semantic importance, significantly reducing the bitrate while preserving critical temporal semantic information. In addition, by integrating trajectory-aligned loss constraints into diffusion processes, we introduce a training-free guidance mechanism in latent space to ensure physically plausible motion patterns without sacrificing the inherent capabilities of generative models. Experimental results demonstrate that T-GVC outperforms both traditional and neural video codecs under ULB conditions. Furthermore, additional experiments confirm that our framework achieves more precise motion control than existing text-guided methods, paving the way for a novel direction of generative video coding guided by geometric motion modeling.

Xianqi Zhang, Hongliang Wei, Wenrui Wang et al.

Humanoid robots have attracted significant attention in recent years. Reinforcement Learning (RL) is one of the main ways to control the whole body of humanoid robots. RL enables agents to complete tasks by learning from environment interactions, guided by task rewards. However, existing RL methods rarely explicitly consider the impact of body stability on humanoid locomotion and manipulation. Achieving high performance in whole-body control remains a challenge for RL methods that rely solely on task rewards. In this paper, we propose a Foundation model-based method for humanoid Locomotion And Manipulation (FLAM for short). FLAM integrates a stabilizing reward function with a basic policy. The stabilizing reward function is designed to encourage the robot to learn stable postures, thereby accelerating the learning process and facilitating task completion. Specifically, the robot pose is first mapped to the 3D virtual human model. Then, the human pose is stabilized and reconstructed through a human motion reconstruction model. Finally, the pose before and after reconstruction is used to compute the stabilizing reward. By combining this stabilizing reward with the task reward, FLAM effectively guides policy learning. Experimental results on a humanoid robot benchmark demonstrate that FLAM outperforms state-of-the-art RL methods, highlighting its effectiveness in improving stability and overall performance.

Chen Liu, Hengyu Man, Xiaopeng Fan et al.

Multi-task dense prediction aims to perform multiple pixel-level tasks simultaneously. However, capturing global cross-task interactions remains non-trivial due to the quadratic complexity of standard self-attention on high-resolution features. To address this limitation, we propose a Multi-Task Linear Semantic Interaction Network (MTLSI-Net), which facilitates cross-task interaction through linear attention. Specifically, MTLSI-Net incorporates three key components: a Multi-Task Multi-scale Query Linear Fusion Block, which captures cross-task dependencies across multiple scales with linear complexity using a shared global context matrix; a Semantic Token Distiller that compresses redundant features into compact semantic tokens, distilling essential cross-task knowledge; and a Cross-Window Integrated attention Block that injects global semantics into local features via a dual-branch architecture, preserving both global consistency and spatial precision. These components collectively enable the network to capture comprehensive cross-task interactions at linear complexity with reduced parameters. Extensive experiments on NYUDv2 and PASCAL-Context demonstrate that MTLSI-Net achieves state-of-the-art performance, validating its effectiveness and efficiency in multi-task learning.

Xiang Tang, Ruotong Li, Xiaopeng Fan

In the field of 3D content generation, single image scene reconstruction methods still struggle to simultaneously ensure the quality of individual assets and the coherence of the overall scene in complex environments, while texture editing techniques often fail to maintain both local continuity and multi-view consistency. In this paper, we propose a novel system ZeroScene, which leverages the prior knowledge of large vision models to accomplish both single image-to-3D scene reconstruction and texture editing in a zero-shot manner. ZeroScene extracts object-level 2D segmentation and depth information from input images to infer spatial relationships within the scene. It then jointly optimizes 3D and 2D projection losses of the point cloud to update object poses for precise scene alignment, ultimately constructing a coherent and complete 3D scene that encompasses both foreground and background. Moreover, ZeroScene supports texture editing of objects in the scene. By imposing constraints on the diffusion model and introducing a mask-guided progressive image generation strategy, we effectively maintain texture consistency across multiple viewpoints and further enhance the realism of rendered results through Physically Based Rendering (PBR) material estimation. Experimental results demonstrate that our framework not only ensures the geometric and appearance accuracy of generated assets, but also faithfully reconstructs scene layouts and produces highly detailed textures that closely align with text prompts.

Xuan Deng, Xingtao Wang, Xiandong Meng et al.

Efficient dynamic point cloud compression (DPCC) critically depends on accurate motion estimation and compensation. However, the inherently irregular structure and substantial local variations of point clouds make this task highly challenging. Existing approaches typically rely on explicit motion estimation, whose encoded motion vectors often fail to capture complex dynamics and inadequately exploit temporal correlations. To address these limitations, we propose a Bidirectional Feature-aligned Motion Transformation (Bi-FMT) framework that implicitly models motion in the feature space. Bi-FMT aligns features across both past and future frames to produce temporally consistent latent representations, which serve as predictive context in a conditional coding pipeline, forming a unified ``Motion + Conditional'' representation. Built upon this bidirectional feature alignment, we introduce a Cross-Transformer Refinement module (CTR) at the decoder side to adaptively refine locally aligned features. By modeling cross-frame dependencies with vector attention, CRT enhances local consistency and restores fine-grained spatial details that are often lost during motion alignment. Moreover, we design a Random Access (RA) reference strategy that treats the bidirectionally aligned features as conditional context, enabling frame-level parallel compression and eliminating the sequential encoding. Extensive experiments demonstrate that Bi-FMT surpasses D-DPCC and AdaDPCC in both compression efficiency and runtime, achieving BD-Rate reductions of 20% (D1) and 9.4% (D1), respectively.

Xuan Deng, Xingtao Wang, Xiandong Meng et al.

In point cloud compression, the quality of a reconstructed point cloud relies on both the global structure and the local context, with existing methods usually processing global and local information sequentially and lacking communication between these two types of information. In this paper, we propose a point-voxel interlaced network (PVINet), which captures global structural features and local contextual features in parallel and performs interactions at each scale to enhance feature perception efficiency. Specifically, PVINet contains a voxel-based encoder (Ev) for extracting global structural features and a point-based encoder (Ep) that models local contexts centered at each voxel. Particularly, a novel conditional sparse convolution is introduced, which applies point embeddings to dynamically customize kernels for voxel feature extraction, facilitating feature interactions from Ep to Ev. During decoding, a voxel-based decoder employs conditional sparse convolutions to incorporate point embeddings as guidance to reconstruct the point cloud. Experiments on benchmark datasets show that PVINet delivers competitive performance compared to state-of-the-art methods.

Kui Jiang, Shiyu Liu, Junjun Jiang et al.

Audio-driven talking head generation holds significant potential for film production. While existing 3D methods have advanced motion modeling and content synthesis, they often produce rendering artifacts, such as motion blur, temporal jitter, and local penetration, due to limitations in representing stable, fine-grained motion fields. Through systematic analysis, we reformulate talking head generation into a unified framework comprising three steps: video preprocessing, motion representation, and rendering reconstruction. This framework underpins our proposed M2DAO-Talker, which addresses current limitations via multi-granular motion decoupling and alternating optimization. Specifically, we devise a novel 2D portrait preprocessing pipeline to extract frame-wise deformation control conditions (motion region segmentation masks, and camera parameters) to facilitate motion representation. To ameliorate motion modeling, we elaborate a multi-granular motion decoupling strategy, which independently models non-rigid (oral and facial) and rigid (head) motions for improved reconstruction accuracy. Meanwhile, a motion consistency constraint is developed to ensure head-torso kinematic consistency, thereby mitigating penetration artifacts caused by motion aliasing. In addition, an alternating optimization strategy is designed to iteratively refine facial and oral motion parameters, enabling more realistic video generation. Experiments across multiple datasets show that M2DAO-Talker achieves state-of-the-art performance, with the 2.43 dB PSNR improvement in generation quality and 0.64 gain in user-evaluated video realness versus TalkingGaussian while with 150 FPS inference speed. Our project homepage is https://m2dao-talker.github.io/M2DAO-Talk.github.io.

Qifan Li, Tianyi Liang, Xingtao Wang et al.

Transformer models have recently garnered significant attention in image restoration due to their ability to capture long-range pixel dependencies. However, long-range attention often results in computational overhead without practical necessity, as degradation and context are typically localized. Normalized average attention distance across various degradation datasets shows that middle-range attention is enough for image restoration. Building on this insight, we propose RouteWinFormer, a novel window-based Transformer that models middle-range context for image restoration. RouteWinFormer incorporates Route-Windows Attnetion Module, which dynamically selects relevant nearby windows based on regional similarity for attention aggregation, extending the receptive field to a mid-range size efficiently. In addition, we introduce Multi-Scale Structure Regularization during training, enabling the sub-scale of the U-shaped network to focus on structural information, while the original-scale learns degradation patterns based on generalized image structure priors. Extensive experiments demonstrate that RouteWinFormer outperforms state-of-the-art methods across 9 datasets in various image restoration tasks.

Hantang Li, Qiang Zhu, Xiandong Meng et al.

In practical applications, low-light images are often compressed for efficient storage and transmission. Most existing methods disregard compression artifacts removal or hardly establish a unified framework for joint task enhancement of low-light images with varying compression qualities. To address this problem, we propose a hybrid priors-guided network (HPGN) that enhances compressed low-light images by integrating both compression and illumination priors. Our approach fully utilizes the JPEG quality factor (QF) and DCT quantization matrix to guide the design of efficient plug-and-play modules for joint tasks. Additionally, we employ a random QF generation strategy to guide model training, enabling a single model to enhance low-light images with different compression levels. Experimental results demonstrate the superiority of our proposed method..

Chenlin Zhou, Han Zhang, Zhaokun Zhou et al.

Deep spiking neural networks (SNNs) have drawn much attention in recent years because of their low power consumption, biological rationality and event-driven property. However, state-of-the-art deep SNNs (including Spikformer and Spikingformer) suffer from a critical challenge related to the imprecise gradient backpropagation. This problem arises from the improper design of downsampling modules in these networks, and greatly hampering the overall model performance. In this paper, we propose ConvBN-MaxPooling-LIF (CML), an SNN-optimized downsampling with precise gradient backpropagation. We prove that CML can effectively overcome the imprecision of gradient backpropagation from a theoretical perspective. In addition, we evaluate CML on ImageNet, CIFAR10, CIFAR100, CIFAR10-DVS, DVS128-Gesture datasets, and show state-of-the-art performance on all these datasets with significantly enhanced performances compared with Spikingformer. For instance, our model achieves 77.64 $\%$ on ImageNet, 96.04 $\%$ on CIFAR10, 81.4$\%$ on CIFAR10-DVS, with + 1.79$\%$ on ImageNet, +1.16$\%$ on CIFAR100 compared with Spikingformer.

Mengyang Wang, Zhicong Zhang, Jiahui Li et al.

Multi-task learning (MTL) is an efficient way to improve the performance of related tasks by sharing knowledge. However, most existing MTL networks run on a single end and are not suitable for collaborative intelligence (CI) scenarios. In this work, we propose an MTL network with a deep joint source-channel coding (JSCC) framework, which allows operating under CI scenarios. We first propose a feature fusion based MTL network (FFMNet) for joint object detection and semantic segmentation. Compared with other MTL networks, FFMNet gets higher performance with fewer parameters. Then FFMNet is split into two parts, which run on a mobile device and an edge server respectively. The feature generated by the mobile device is transmitted through the wireless channel to the edge server. To reduce the transmission overhead of the intermediate feature, a deep JSCC network is designed. By combining two networks together, the whole model achieves 512x compression for the intermediate feature and a performance loss within 2% on both tasks. At last, by training with noise, the FFMNet with JSCC is robust to various channel conditions and outperforms the separate source and channel coding scheme.

Hengyu Man, Xiaopeng Fan, Ruiqin Xiong et al.

As a crucial part of video compression, intra prediction utilizes local information of images to eliminate the redundancy in spatial domain. In both the High Efficiency Video Coding (H.265/HEVC) and Versatile Video Coding (H.266/VVC), multiple directional prediction modes are employed to find the texture trend of each small block and then the prediction is made based on reference samples in the selected direction. Recently, the intra prediction schemes based on neural networks have achieved great success. In these methods, the networks are trained and applied to intra prediction to assist the directional prediction modes. In this paper, we propose a novel tree-structured data clustering-driven neural network (dubbed TreeNet) for intra prediction, which builds the networks and clusters the training data in a tree-structured manner. Specifically, in each network split and training process of TreeNet, every parent network on a leaf node is split into two child networks by adding or subtracting Gaussian random noise. Then a data clustering-driven training is applied to train the two derived child networks using the clustered training data of their parent. To test the performance, TreeNet is integrated into VVC and HEVC to combine with or replace the directional prediction modes. In addition, a fast termination strategy is proposed to accelerate the search of TreeNet. The experimental results demonstrate that TreeNet with the fast termination can reach an average of 2.8% Bjontegaard distortion rate (BD-rate) improvement (up to 8.1%) and 4.9% BD-rate improvement (up to 8.2%) over VVC (VTM-4.0) and HEVC (HM-16.9) with all intra configuration, respectively.