cs.CVNov 4, 2026

CoDecide: Content-Adaptive Video Decoding for High-Throughput Video Analytics

Authors: Shristy Shah, Thomas B. Moeslund, Kamal Nasrollahi

Organizations: Aalborg University, Denmark · Milestone Systems, Denmark

Abstract

Analytics on video archives are bottlenecked by decoding. In a dense pipeline that runs a detector on every frame, decoding is a small fraction of the runtime, but once the detector is restricted to selected frames, decoding takes over as the main cost. We introduce CoDecide, a retrospective analytics system that only decodes the frames where new content appears, using signals from the compressed bitstream to find them. Motion tells us something is moving, but not whether it is new to the scene or has been there the whole time. To make that distinction, we supervise a lightweight compressed-domain student on a label-free novelty signal from a frozen DINOv2 teacher, which fires on semantically new regions rather than on any motion. Consequently, the student model generalizes to new camera deployments without requiring per-camera retraining. At inference, it decides which frames to decode and which regions to reconstruct. Detections from the decoded frames are propagated to the skipped frames using motion vectors. We evaluate CoDecide with a leave-one-dataset-out protocol on VIRAT, MEVA, and ShanghaiTech. Against a dense per-frame detector, CoDecide recovers more moving-object detections than prior compressed-domain selectors at an equivalent decoding budget, outperforming even a competitor trained directly on the target camera. CoDecide runs 15 to 24x faster than a pipeline that decodes and detects every frame. Code is available at https://github.com/shristy51/CoDecide.

Explore similar work

CardsList
  1. CodecMotion-OAD: Efficient Online Action Detection with Compressed-Domain Motion

    Oct 17, 2026Zepeng Sun, Naichuan Zheng, Hailun Xia +1Temporal Action DetectionHuman Activity Recognition

  2. ComPrivDet: Efficient Privacy Object Detection in Compressed Domains Through Inference Reuse

    Apr 4, 2026Yunhao Yao, Zhiqiang Wang, Ruiqi Li +3Object DetectionVideo Object Detection

  3. Detect Early, Escalate Rarely: Anytime Detection of AI-Generated Video from the Compressed Bitstream

    Jul 21, 2026Mert Onur Cakiroglu, Mehmet Dalkilic, Hasan KurbanDeepfake DetectionSequential Hypothesis Testing