iQIYI Technical Product Team
Apr 10, 2020 · Big Data
Video Copyright Detection Solution Using SE-ResNeXt and Faiss in the 2019 CCF Big Data & Computational Intelligence Competition
The iQiyi team “都挺好” tackled the 2019 CCF Video Copyright Detection contest by extracting frame‑level features with SE‑ResNeXt, indexing them with Faiss, aligning temporally via a critical‑path method, and refining copy boundaries using SIFT re‑matching and a sliding‑window approach, ultimately achieving an F1 score of 0.9678 after three iterative stages of model selection, cascade detection, and feature fusion.
CCF competitionSE-ResNeXtVideo Copyright Detection
0 likes · 6 min read