The paper introduces a neural video compression approach that combines deformable temporal alignment with difference-aware spatial selective fusion. A Context-aware Temporal Alignment Module generates complementary temporal context, while a Difference-aware Spatial Selective Fusion module adaptively selects reliable temporal information and suppresses misalignment. Experiments demonstrate that this method improves rate-distortion performance compared to DCVC-DC.
By Chuyue Shan, Songlin Sun, Wang Chenwei, Shen Zihan
The paper introduces Neural Image Compression with Internal-External Prior Fusion and Alignment (IE-PFA), a context model that combines internal priors from the current image with an external dictionary shared by encoder and decoder. It employs a RWKV-based Spatial-Channel Prior Fusion (RSCPF) module for efficient fusion and a Self-Distilled Prior Alignment (SDPA) mechanism for alignment during training. The model supports variable-rate coding across 64 quality levels, achieves a 21.65% BD-Rate saving over VTM-22.0-Intra, uses only 29.9M parameters, and offers fast encoding (61 ms) and decoding (52 ms) speeds.
By Tiancheng Si, Wenlong Xu, Dahua Gao
TCNeRV is a new implicit neural video compression method that models temporal context in both feature and embedding domains. Its multi‑scale temporal‑context fusion module injects gated historical features across decoder scales, while temporal embedding‑residual coding predicts and encodes only the residual of each content embedding. With about 3 million parameters, TCNeRV achieves an average PSNR of 36.08 dB on the UVG dataset, outperforming HNeRV‑Boost by 2.20 dB and reducing BD‑rate by 22.06%, 66.73%, and 29.85% relative to HM, DCVC, and HiNeRV respectively.
By Xuezhi Xiang, Yixin Zhao, Heqi Xiang, Jiayao Liu, Shanjun Zhang
In video understanding, vision-language models (VLMs) must ingest massive numbers of visual tokens, causing the computational and memory cost of the prefill stage to rise sharply. Such visual sequences are highly redundant along the spatio-temporal dimension, yet a high compression ratio is often accompanied by the loss of critical details.
arXiv:2610.34765v1 Announce Type: new
Abstract: Multi-modal video fusion aims to generate temporally coherent videos with salient targets and rich structural details by integrating complementary info...
By Jingxue Huang, Xiaosong Li, Yi Lin
The paper introduces Event-guided Neural Video Codec (ENVC), a neural video compression method that incorporates event streams—capturing brightness changes between frames—into both motion and frame coding stages. By using event-guided motion priors and event-conditioned predictors, ENVC improves RGB compression efficiency, achieving significant BD-rate savings across six benchmarks. The authors also synthesize paired RGB-event data for training and demonstrate that the gains persist on large-motion sequences, highlighting events as a valuable complementary modality for video coding.
By Jiyun Kong, Jungwoo Kim, Enes Eray Demirtas, Touradj Ebrahimi, Jong-Seok Lee
arXiv:2603.17546v2 Announce Type: replace
Abstract: Perceptual video compression leverages generative priors to reconstruct realistic textures and motions at low bitrates. However, existing perceptua...
By Daowen Li, Ruixiao Dong, Kai Li, Ying Chen, Ding Ding, Li Li
arXiv:2605. 20301v2 Announce Type: replace-cross Abstract: In autonomous driving, 3D object detection is essential for accurate perception and reliable decision-making.
By Wenxuan Li, Qin Zou, Shoubing Chen, Chi Chen, Yingyi Yang, Qingxiang Meng
Video large language models (VLLMs) need long visual token sequences, making token compression essential for efficient inference. This paper introduces a training‑free token compression technique that leverages existing video codec information—such as reference‑frame dependencies, motion vectors, and residuals—to guide low‑cost search priors. By aligning these codec cues with the visual token grid, the method creates compact candidate pools for token protection and localized candidates for temporal token merging, reducing feature comparisons while maintaining performance comparable to or better than the FlashVID baseline across multiple VLLMs and benchmarks.
arXiv:2606. 13580v1 Announce Type: cross Abstract: Event-based vision has drawn increasing attention owing to its distinctive properties, including ultra-high temporal resolution and extreme dynamic range.
By Dachun Kai, Jiayao Lu, Yueyi Zhang, Xiaoyan Sun
The paper introduces S2A, a semantic-to-spatial alignment framework designed for alignment‑free RGB‑T salient object detection. It employs a global‑guided hierarchical fusion module to refine intra‑modal features, an alignment‑free cross‑modal channel attention module to exchange semantic information, and a spatial deformable cross‑attention module to recover local spatial correspondence. These components collectively reduce misalignment‑induced feature contamination and achieve competitive performance on public benchmarks without additional bells and whistles.
By Qiangqiang Zhou, Yang Luo, Yong Chen, Jiawei Xu
Video large language models (VLLMs) need long visual token sequences, making token compression essential for efficient inference. This paper introduces a training‑free token compression method that leverages existing video codec information—such as reference‑frame dependencies, motion vectors, and residuals—to guide low‑cost search for token protection and temporal merging. Experiments on three VLLMs and four benchmarks show that the method matches or surpasses the FlashVID baseline while dramatically reducing feature comparisons.
By Xihua Sheng, Ruijie Liu, Chang Wen Chen