CodecSight: Leveraging Video Codec Signals for Efficient Streaming VLM Inference

• URL: http://arxiv.org/abs/2604.06036v3
• Date: Thu, 09 Apr 2026 09:40:36 GMT
• Title: CodecSight: Leveraging Video Codec Signals for Efficient Streaming VLM Inference
• Abstract summary: We present CodecSight, a streaming video analytics system built on a key observation that video codecs already extract the temporal and spatial structure of each stream as a byproduct of compression.<n>CodecSight treats this metadata as a low-cost signal to unify runtime optimization across video decoding, visual processing, and LLM prefilling, with transmission reduction as an inherent benefit of operating directly on compressed bitstreams.<n> Experiments show that CodecSight achieves an improvement in throughput of up to 3$times$, and a reduction of up to 87% in GPU compute over state-of-the-art baselines, maintaining competitive accuracy with
• Score: 4.835489391255295
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Video streaming analytics is a crucial workload for vision-language model serving, but the high cost of multimodal inference limits scalability. Prior systems reduce inference cost by exploiting temporal and spatial redundancy in video streams, but they target either the vision transformer (ViT) or the LLM with a limited view, leaving end-to-end opportunities untapped. Moreover, existing methods incur significant overhead to identify redundancy, either through offline profiling and training or costly online computation, making them ill-suited for dynamic real-time streams. We present CodecSight, a codec-guided streaming video analytics system, built on a key observation that video codecs already extract the temporal and spatial structure of each stream as a byproduct of compression. CodecSight treats this codec metadata as a low-cost runtime signal to unify optimization across video decoding, visual processing, and LLM prefilling, with transmission reduction as an inherent benefit of operating directly on compressed bitstreams. This drives codec-guided patch pruning before ViT encoding and selective key-value cache refresh during LLM prefilling, both of which are fully online and do not require offline training. Experiments show that CodecSight achieves an improvement in throughput of up to 3$\times$, and a reduction of up to 87% in GPU compute over state-of-the-art baselines, maintaining competitive accuracy with only 0$\sim$8% F1 drop.

Related papers

• LiteFrame: Efficient Vision Encoders Unlock Frame Scaling in Video LLMs [90.77662862634509]
  LiteFrame is a strong, yet highly efficient backbone for Video Large Language Models.<n>LiteFrame provides a 35% reduction in end-to-end latency while processing 8$times$ more frames.<n>Our results demonstrate a potential path to unlocking longer-form video understanding under fixed compute budgets.
  arXiv  Detail & Related papers  (2026-05-17T05:02:52Z)
• CoPE-VideoLM: Codec Primitives For Efficient Video Language Models [56.76440182038839]
  Video Language Models (VideoLMs) empower AI systems to understand temporal dynamics in videos.<n>Current methods use sampling which can miss both macro-level events and micro-level details due to the sparse temporal coverage.<n>We propose to leverage video primitives which encode video redundancy and sparsity without requiring expensive full-image encoding for most frames.
  arXiv  Detail & Related papers  (2026-02-13T18:57:31Z)
• V-Rex: Real-Time Streaming Video LLM Acceleration via Dynamic KV Cache Retrieval [1.677021230191566]
  Streaming video large language models (LLMs) are increasingly used for real-time multimodal tasks such as video captioning, question answering, conversational agents, and augmented reality.<n>These models face fundamental memory and computational challenges because their key-value ( KV) caches grow substantially with continuous streaming video input.<n>We propose V-Rex, the first software- hardware co-designed accelerator that addresses both algorithmic and hardware bottlenecks in streaming video LLM inference.
  arXiv  Detail & Related papers  (2025-12-13T11:02:04Z)
• FocalCodec-Stream: Streaming Low-Bitrate Speech Coding via Causal Distillation [27.32235541083431]
  FocalCodec-Stream is a hybrid that compresses speech into a single binary codebook at 0.55 - 0.80 kbps with a theoretical latency of 80 ms.<n> Experiments show that FocalCodec-Stream outperforms existing streamable codecs at comparables.
  arXiv  Detail & Related papers  (2025-09-19T17:57:13Z)
• SIEDD: Shared-Implicit Encoder with Discrete Decoders [36.705337163276255]
  Implicit Neural Representations (INRs) offer exceptional fidelity for video compression by learning per-video optimized functions.<n>Existing attempts to accelerate INR encoding often sacrifice reconstruction quality or crucial coordinate-level control.<n>We introduce SIEDD, a novel architecture that fundamentally accelerates INR encoding without these compromises.
  arXiv  Detail & Related papers  (2025-06-29T19:39:43Z)
• LiveVLM: Efficient Online Video Understanding via Streaming-Oriented KV Cache and Retrieval [13.891391928767195]
  LiveVLM is a training-free framework specifically designed for streaming, online video understanding and real-time interaction.<n>LiveVLM constructs a streaming-oriented KV cache to process video streams in real-time, retain long-term video details and eliminate redundant KVs.<n>When a new question is proposed, LiveVLM incorporates an online question-answering process that efficiently fetches both short-term and long-term visual information.
  arXiv  Detail & Related papers  (2025-05-21T08:47:15Z)
• Plug-and-Play Versatile Compressed Video Enhancement [57.62582951699999]
  Video compression effectively reduces the size of files, making it possible for real-time cloud computing.<n>However, it comes at the cost of visual quality, challenges the robustness of downstream vision models.<n>We present a versatile-aware enhancement framework that adaptively enhance videos under different compression settings.
  arXiv  Detail & Related papers  (2025-04-21T18:39:31Z)
• VideoScan: Enabling Efficient Streaming Video Understanding via Frame-level Semantic Carriers [23.541896057977745]
  VideoScan is an efficient vision-language model (VLM) inference framework for real-time video interaction.<n>VideoScan employs a single semantic carrier token to represent each frame.
  arXiv  Detail & Related papers  (2025-03-12T13:30:40Z)
• STORM: Token-Efficient Long Video Understanding for Multimodal LLMs [116.4479155699528]
  STORM is a novel architecture incorporating a dedicated temporal encoder between the image encoder and the Video-LLMs.<n>We show that STORM achieves state-of-the-art results across various long video understanding benchmarks.
  arXiv  Detail & Related papers  (2025-03-06T06:17:38Z)
• Low-Latency Neural Stereo Streaming [6.49558286032794]
  Low-Latency neural for Stereo video Streaming (LLSS) is a novel parallel stereo video coding method designed for low-latency stereo video streaming. LLSS processes left and right views in parallel, minimizing latency; all while substantially improving R-D performance compared to both existing neural and conventional codecs.
  arXiv  Detail & Related papers  (2024-03-26T17:11:51Z)
• A Coding Framework and Benchmark towards Low-Bitrate Video Understanding [63.05385140193666]
  We propose a traditional-neural mixed coding framework that takes advantage of both traditional codecs and neural networks (NNs) The framework is optimized by ensuring that a transportation-efficient semantic representation of the video is preserved. We build a low-bitrate video understanding benchmark with three downstream tasks on eight datasets, demonstrating the notable superiority of our approach.
  arXiv  Detail & Related papers  (2022-02-06T16:29:15Z)
• Low-Fidelity End-to-End Video Encoder Pre-training for Temporal Action Localization [96.73647162960842]
  TAL is a fundamental yet challenging task in video understanding. Existing TAL methods rely on pre-training a video encoder through action classification supervision. We introduce a novel low-fidelity end-to-end (LoFi) video encoder pre-training method.
  arXiv  Detail & Related papers  (2021-03-28T22:18:14Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.