Papers for

video streaming providers

Papers whose findings have a practical use for this group, as judged from the abstract. Open a paper to read what it means in practice.

Bitstream native restorer improves corrupted image recovery accuracy

BiNar: Bitstream Native Restorer for Bitstream-corrupted Image Restoration

Abstract: Bitstream-corrupted image restoration (BCIR) aims to reconstruct faithful images from corrupted compressed bitstreams. However, corruption occurs in bitstream-domain may lead to unpredictable error propagation and distortion amplification in pixel-domain, and even result in complete undecodability by standard decoders, posing significant challenge to existing image restoration paradigms. To address this challenge, we propose Bitstream Native Restorer (BiNar), a generative restoration framework that leverages fine-grained bitstream-native semantics to guide bitstream-corrupted image restoration. In BiNar, Bitstream Semantic Modeling (BSeM) extracts robust informative semantics and models sequential bytes directly from corrupted bitstream, enabling to generate fine-grained bitstream descriptions without relying on pixel decoding. Based on the descriptions,the Corruption-adaptive Multimodal Semantic Aggregation (CMSA) further extracts adaptive visual modal cues in decoded corrupted images and aggregates to bitstream descriptions. The multimodal aggregated fine-grained descriptions can serve as coherent restoration guidance. Finally, Blind Corruption Localization (BCoL) exploits the segmentation prior of visual foundation models to identify and locate error regions that requires recovery in corrupted images. The located error regions and the fine-grained descriptions jointly serve as visual prompt and language prompt of diffusion models image restoration, respectively. Extensive experiments under diverse bitstream corruption conditions show that BiNar outperforms all State-of-the-arts image restoration models regardless of perceptual quality or objective quality. Ablation studies demonstrate the effectiveness of fine-grained bitstream-native semantics in BCIR task.

Fri 16 OctComputer Vision and Pattern Recognition
The gist
Sometimes images get damaged because their compressed data files have errors that normal decoders can't fix, leading to blurry or broken pictures. The authors created BiNar, a tool that works directly with the corrupted data to understand its structure before trying to fix the image. It combines this deep data insight with visual clues from the partly corrupted images to better locate and repair the damaged parts. This approach allows their method to restore images more accurately than existing techniques, even when the data is badly corrupted.
Open → 2610.24094v1

Real-time video compression using deformable 2D Gaussian splatting

Deformable 2D Gaussian Splatting for Efficient 4K Video Compression

Abstract: Ultra-High-Definition (UHD) video presents significant challenges for efficient storage and real-time decoding. Learning-based methods, such as Neural Video Compression (NVC) and Implicit Neural Representations (INR), achieve competitive rate-distortion performance but suffer from high decoding latency and excessive memory usage. Meanwhile, Gaussian Splatting has recently attracted attention in the computer graphics community due to its ultra-fast rendering and high-fidelity visual quality. Despite these advantages, its application in video compression remains largely unexplored. To bridge this gap, we propose a real-time video compression framework that represents and compresses a Group of Pictures (GOP) using a coarse-to-fine multi-scale 2D Gaussian Splatting (2DGS) structure coupled with a lightweight deformation network. Experiments demonstrate that our method delivers rate-distortion performance in LPIPS that surpasses H.265 and other state-of-the-art learning-based video compression methods. Our work demonstrates the potential of Gaussian Splatting as a practical solution for efficient high-resolution video compression.

Sat 12 SeptComputer Vision and Pattern Recognition
The gist
Ultra-high-definition videos require a lot of storage and are slow to decode in real time. The authors propose a new way to compress videos by using a 2D Gaussian Splatting method combined with a lightweight deformation network. This approach can compress groups of video frames quickly and efficiently, resulting in better video quality compared to the popular H.265 standard and other recent AI-based methods. Their method is especially suited for very high-resolution videos, like 4K.
Open → 2609.14129v1