1st place solution for AVA-Kinetics Crossover in AcitivityNet Challenge 2020
arXiv:2006.09116
Abstract
This technical report introduces our winning solution to the spatio-temporal action localization track, AVA-Kinetics Crossover, in ActivityNet Challenge 2020. Our entry is mainly based on Actor-Context-Actor Relation Network. We describe technical details for the new AVA-Kinetics dataset, together with some experimental results. Without any bells and whistles, we achieved 39.62 mAP on the test set of AVA-Kinetics, which outperforms other entries by a large margin. Code will be available at: https://github.com/Siyu-C/ACAR-Net.
arXiv commentarXiv admin note: substantial text overlap with arXiv:2006.07976
References in corpus (8)
- Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks
- Accurate, Large Minibatch SGD: Training ImageNet in 1 Hour
- The Open Images Dataset V4: Unified image classification, object detection, and visual relationship detection at scale
- A Short Note on the Kinetics-700-2020 Human Action Dataset
- The AVA-Kinetics Localized Human Actions Video Dataset
- PFDet: 2nd Place Solution to Open Images Challenge 2018 Object Detection Track
- Actor-Context-Actor Relation Network for Spatio-Temporal Action Localization
- Three Branches: Detecting Actions With Richer Features