Towards Improving Spatiotemporal Action Recognition in Videos

December 15, 2020 · Entered Twilight · 🏛 arXiv.org

"Last commit was 5.0 years ago (≥5 year threshold)"

Evidence collected by the PWNC Scanner

Repo contents: .DS_Store, .gitignore, BalancedCrossEntropy.py, EffectiveNumberofObjectClass.py, FocalLoss.py, InverseClassFrequency.py, README.md, backbones_2d, backbones_3d, cfam.py, cfg.py, cfg, clip.py, darknet.py, data, dataset.py, datasets, detections, eval_results.py, evaluation, examples, finalAnnots.mat, linknet.py, model.py, opts.py, region_loss.py, run_agot-24.sh, run_frame_mAP_agot.sh, run_frame_mAP_jhmdb.sh, run_frame_mAP_ucf.sh, run_jhmdb-21.sh, run_ucf101-24.sh, run_video_mAP_jhmdb.sh, run_video_mAP_ucf.sh, train.py, utils.py, video_mAP.py, weights

Authors Shentong Mo, Xiaoqing Tan, Jingfei Xia, Pinxu Ren arXiv ID 2012.08097 Category cs.CV: Computer Vision Citations 2 Venue arXiv.org Repository https://github.com/stoneMo/YOWOv2 ⭐ 9 Last Checked 2 months ago

Abstract

Spatiotemporal action recognition deals with locating and classifying actions in videos. Motivated by the latest state-of-the-art real-time object detector You Only Watch Once (YOWO), we aim to modify its structure to increase action detection precision and reduce computational time. Specifically, we propose four novel approaches in attempts to improve YOWO and address the imbalanced class issue in videos by modifying the loss function. We consider two moderate-sized datasets to apply our modification of YOWO - the popular Joint-annotated Human Motion Data Base (J-HMDB-21) and a private dataset of restaurant video footage provided by a Carnegie Mellon University-based startup, Agot.AI. The latter involves fast-moving actions with small objects as well as unbalanced data classes, making the task of action localization more challenging. We implement our proposed methods in the GitHub repository https://github.com/stoneMo/YOWOv2.