Jun 28, 2019 · Artificial Intelligence

Watchdog Team's TOP1 Solution for the iQIYI & ACMMM2019 Multimodal Video Person Recognition Challenge

Watchdog team won TOP1 in iQIYI & ACMMM2019 multimodal video person recognition challenge using pre‑extracted multimodal features, a 2048‑dim classifier with BCE loss, re‑ranking, DALI‑accelerated re‑detection, fine‑tuned InsightFace, and multi‑model ensembling achieving ~91% test accuracy.

Multimodal Learningfeature fusionmodel ensemble

0 likes · 12 min read

Watchdog Team's TOP1 Solution for the iQIYI & ACMMM2019 Multimodal Video Person Recognition Challenge

iQIYI Technical Product Team

Mar 22, 2019 · Artificial Intelligence

Experience Report of the 2018 iQIYI Multimodal Video Person Identification Challenge (WitcheR Team)

The WitcheR team won the 2018 iQIYI multimodal video person identification challenge by building a fast pipeline that combined a custom face‑and‑keypoint detector, ArcFace‑trained face embeddings, scene classification, and a three‑layer MLP with several training tricks, achieving a final mAP of 88.6 % and demonstrating the value of rapid idea validation and open‑sourced code for future challenges.

MLPModel FusionMultimodal

0 likes · 12 min read

Experience Report of the 2018 iQIYI Multimodal Video Person Identification Challenge (WitcheR Team)