12 papers
CoViST: Visual Token Compression via Composable States
Qi Zhang, Xiandong Meng, Ronggang Wang +1
Visual token compression lowers the inference cost of vision--language models by representing images with fewer tokens. However, most existing methods compress visual tokens to a r…
CLEAR: Conflict-aware Learning via Evidence-guided Adaptive Routing for Unified Sparse-View 3D Gaussian Super-Resolution
Hantang Li, Qiang Zhu, Xiandong Meng +2
Sparse-view 3D Gaussian Splatting Super-resolution is highly challenging since the sparse and low-resolution (LR) inputs lack sufficient geometric and high-frequency information fo…
LAST: The Last Query Token Guides Visual Token Pruning for Edge-Cloud Collaborative MLLM Inference
Feng Yang, Xinrui Ju, Keyang Zhang +6
Multimodal foundation models are reshaping edge-cloud visual intelligence from task-specific feature pipelines into token-based interfaces, where edge devices encode visual inputs…
LUMI: Tokenizer-Agnostic LLM-Based Lossless Image Compression
Chris Xing Tian, Chengkai Wu, Ziyu Wang +6
Large language model (LLM)-based lossless image compression methods typically represent pixel data through the native text interface of a pretrained model, converting pixel values…
Neural Video Compression with Domain Transfer
Tiange Zhang, Rongqun Lin, Xiandong Meng +4
Content-adaptive compression has always been a key direction in neural video coding (NVC), aiming to mitigate the domain gap between training and testing data. Such gaps often aris…
PairDropGS: Paired Dropout-Induced Consistency Regularization for Sparse-View Gaussian Splatting
Hantang Li, Qiang Zhu, Xiandong Meng +3
Dropout-based sparse-view 3D Gaussian Splatting (3DGS) methods alleviate overfitting by randomly suppressing Gaussian primitives during training. Existing methods mainly focus on d…
NTIRE 2026 3D Restoration and Reconstruction in Real-world Adverse Conditions: RealX3D Challenge Results
Shuhong Liu, Chenyu Bao, Ziteng Cui +103
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…
DOC-GS: Dual-Domain Observation and Calibration for Reliable Sparse-View Gaussian Splatting
Hantang Li, Qiang Zhu, Xiandong Meng +2
Sparse-view reconstruction with 3D Gaussian Splatting (3DGS) is fundamentally ill-posed due to insufficient geometric supervision, often leading to severe overfitting and the emerg…
LG-HCC: Local Geometry-Aware Hierarchical Context Compression for 3D Gaussian Splatting
Xuan Deng, Xiandong Meng, Hengyu Man +4
Although 3D Gaussian Splatting (3DGS) enables high-fidelity real-time rendering, its prohibitive storage overhead severely hinders practical deployment. Recent anchor-based 3DGS co…
Bidirectional Feature-aligned Motion Transformation for Efficient Dynamic Point Cloud Compression
Xuan Deng, Xingtao Wang, Xiandong Meng +4
Efficient dynamic point cloud compression (DPCC) critically depends on accurate motion estimation and compensation. However, the inherently irregular structure and substantial loca…
PVINet: Point-Voxel Interlaced Network for Point Cloud Compression
Xuan Deng, Xingtao Wang, Xiandong Meng +2
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 an…
Predicting Satisfied User and Machine Ratio for Compressed Images: A Unified Approach
Qi Zhang, Shanshe Wang, Xinfeng Zhang +3
Nowadays, high-quality images are pursued by both humans for better viewing experience and by machines for more accurate visual analysis. However, images are usually compressed bef…