1 code implementation • IEEE Access 2023 • Si-Dong Roh, Ki-Seok Chung
To effectively refine the box from the degraded images in the videos, we used three novel approaches: cascade refinement, dynamic core-set conditioning, and local batch refinement.
Ranked #1 on Video Object Detection on ImageNet VID
no code implementations • IEEE Access 2022 • Si-Dong Roh, Ki-Seok Chung
Our method with global and local attention stages obtains 84. 5 and 85. 9 mAP on ResNet-101 and ResNeXt-101, respectively, thus achieving state-of-the-art performance without requiring additional post-processing methods.
Ranked #10 on Video Object Detection on ImageNet VID