Beyond Pixels: Mining Compressed Domain Artifacts for Efficient AI-Generated Video Detection

Anran Zhu, Zhengli Shi, Chende Zheng, Chenhao Lin, Zhengyu Zhao, Le Yang, Chong Zhang, Shuai Liu, Chao Shen
Proceedings of the 43rd International Conference on Machine Learning, PMLR 306:167689-167701, 2026.

Abstract

With the rapid advancement of high-fidelity video generation models, robust AI-generated video (AIGV) detection has become increasingly needed. While most AIGV detection methods operate in the decoded pixel domain, we observe that detection in the pixel domain inevitably entangles task-irrelevant semantic information, leading to substantial semantic redundancy and extensive redundant computation, while overlooking free-to-use signals in compressed bitstreams. In particular, motion vectors and residuals directly encode temporal and spatial generative artifacts but remain largely underexplored. To address these issues, we propose a unified framework for Spatio-Temporal REsidual and Artifact Mining, namely STREAM, which enables AIGV detection directly from compressed bitstreams. STREAM leverages I-frames, motion vectors, and residual errors to capture spatiotemporal artifacts that are typically smoothed out by decompression filters. In particular, we design a lightweight network with a motion-guided alignment module and a gated fusion mechanism, enabling adaptive fusion of spatial artifacts and nonlinear temporal dynamics. Extensive experimental results demonstrate that STREAM achieves SOTA performance with an mAP of 0.965, with 2.5$\times$ faster inference than previous SOTA baselines.

Cite this Paper


BibTeX
@InProceedings{pmlr-v306-zhu26ap, title = {Beyond Pixels: Mining Compressed Domain Artifacts for Efficient {AI}-Generated Video Detection}, author = {Zhu, Anran and Shi, Zhengli and Zheng, Chende and Lin, Chenhao and Zhao, Zhengyu and Yang, Le and Zhang, Chong and Liu, Shuai and Shen, Chao}, booktitle = {Proceedings of the 43rd International Conference on Machine Learning}, pages = {167689--167701}, year = {2026}, editor = {Zhang, Tong and Dudik, Miroslav and Jaggi, Martin and Agarwal, Alekh and Li, Sharon and Schuurmans, Dale and Zhu, Jerry and Berkenkamp, Felix and Dong, Hanze and Bietti, Alberto}, volume = {306}, series = {Proceedings of Machine Learning Research}, month = {06--11 Jul}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v306/main/assets/zhu26ap/zhu26ap.pdf}, url = {https://proceedings.mlr.press/v306/zhu26ap.html}, abstract = {With the rapid advancement of high-fidelity video generation models, robust AI-generated video (AIGV) detection has become increasingly needed. While most AIGV detection methods operate in the decoded pixel domain, we observe that detection in the pixel domain inevitably entangles task-irrelevant semantic information, leading to substantial semantic redundancy and extensive redundant computation, while overlooking free-to-use signals in compressed bitstreams. In particular, motion vectors and residuals directly encode temporal and spatial generative artifacts but remain largely underexplored. To address these issues, we propose a unified framework for Spatio-Temporal REsidual and Artifact Mining, namely STREAM, which enables AIGV detection directly from compressed bitstreams. STREAM leverages I-frames, motion vectors, and residual errors to capture spatiotemporal artifacts that are typically smoothed out by decompression filters. In particular, we design a lightweight network with a motion-guided alignment module and a gated fusion mechanism, enabling adaptive fusion of spatial artifacts and nonlinear temporal dynamics. Extensive experimental results demonstrate that STREAM achieves SOTA performance with an mAP of 0.965, with 2.5$\times$ faster inference than previous SOTA baselines.} }
Endnote
%0 Conference Paper %T Beyond Pixels: Mining Compressed Domain Artifacts for Efficient AI-Generated Video Detection %A Anran Zhu %A Zhengli Shi %A Chende Zheng %A Chenhao Lin %A Zhengyu Zhao %A Le Yang %A Chong Zhang %A Shuai Liu %A Chao Shen %B Proceedings of the 43rd International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2026 %E Tong Zhang %E Miroslav Dudik %E Martin Jaggi %E Alekh Agarwal %E Sharon Li %E Dale Schuurmans %E Jerry Zhu %E Felix Berkenkamp %E Hanze Dong %E Alberto Bietti %F pmlr-v306-zhu26ap %I PMLR %P 167689--167701 %U https://proceedings.mlr.press/v306/zhu26ap.html %V 306 %X With the rapid advancement of high-fidelity video generation models, robust AI-generated video (AIGV) detection has become increasingly needed. While most AIGV detection methods operate in the decoded pixel domain, we observe that detection in the pixel domain inevitably entangles task-irrelevant semantic information, leading to substantial semantic redundancy and extensive redundant computation, while overlooking free-to-use signals in compressed bitstreams. In particular, motion vectors and residuals directly encode temporal and spatial generative artifacts but remain largely underexplored. To address these issues, we propose a unified framework for Spatio-Temporal REsidual and Artifact Mining, namely STREAM, which enables AIGV detection directly from compressed bitstreams. STREAM leverages I-frames, motion vectors, and residual errors to capture spatiotemporal artifacts that are typically smoothed out by decompression filters. In particular, we design a lightweight network with a motion-guided alignment module and a gated fusion mechanism, enabling adaptive fusion of spatial artifacts and nonlinear temporal dynamics. Extensive experimental results demonstrate that STREAM achieves SOTA performance with an mAP of 0.965, with 2.5$\times$ faster inference than previous SOTA baselines.
APA
Zhu, A., Shi, Z., Zheng, C., Lin, C., Zhao, Z., Yang, L., Zhang, C., Liu, S. & Shen, C.. (2026). Beyond Pixels: Mining Compressed Domain Artifacts for Efficient AI-Generated Video Detection. Proceedings of the 43rd International Conference on Machine Learning, in Proceedings of Machine Learning Research 306:167689-167701 Available from https://proceedings.mlr.press/v306/zhu26ap.html.

Related Material