Abstract:This notebook paper presents an overview and comparative analysis of our systems designed for the following three tasks in ActivityNet Challenge 2017: trimmed action recognition, temporal action proposals and densecaptioning events in videos. Trimmed Action Recognition (TAR): We investigate and exploit multiple spatio-temporal clues for trimmed action recognition (TAR) task, i.e., frame, short video clip and motion (optical flow) by leveraging 2D or 3D convolutional neural networks (CNNs). The mechanism of different quantization methods is studied as well. Furthermore, improved dense trajectory with fisher vector encoding over the whole trimmed video is utilized. All activities are finally classified by late fusing the predictions of one-versus-rest linear SVMs learnt on each clue. Temporal Action Proposals (TAP): To generate temporal action proposals from videos, a three-stage workflow is particularly devised for TAP task. Given an untrimmed video, our system firstly generates an actionness curve via a snippet-level actionness classifier. The temporal actionness grouping scheme is then exploited over actionness curve to produce proposal candidates. Finally, a proposal re-ranking procedure is incorporated to select high-quality proposals via a proposal-level actionness classifier. Dense-Captioning Events in Videos (DCEV): For DCEV task, we firstly adopt our temporal action proposal system mentioned above to localize temporal proposals of interest in video, and then generate the descriptions for each proposal. Specifically, RNNs encode a given video and its detected attributes into a fixed dimensional vector, and then decode it to the target output sentence. Moreover, we extend the attributes-based CNNs plus RNNs model with policy gradient optimization and retrieval mechanism to further boost video captioning performance.

Technical Report for SoccerNet Challenge 2022 – Replay Grounding Task

Technical Report for ActivityNet Challenge 2022 – Temporal Action Localization

SoccerNet 2022 Challenges Results.

SoccerNet 2024 Challenges Results

Automatic replay generation for soccer video broadcasting.

SoccerNet 2023 Challenges Results

Complementary Boundary Generator with Scale-Invariant Relation Modeling for Temporal Action Localization: Submission to ActivityNet Challenge 2020

Replay Scene Classification in Soccer Video Using Web Broadcast Text

CBR-Net: Cascade Boundary Refinement Network for Action Detection: Submission to ActivityNet Challenge 2020 (task 1)

Feature Combination Meets Attention: Baidu Soccer Embeddings and Transformer based Temporal Detection

CUHK ETHZ SIAT Submission to ActivityNet Challenge 2016

A Transformer-based System for Action Spotting in Soccer Videos

Temporal Fusion Network for Temporal Action Localization:Submission to ActivityNet Challenge 2020 (Task E)

A statistical framework for replay detection in soccer video

Multi-camera Temporal Grouping for Play/Break Event Detection in Soccer Games

SoccerNet Game State Reconstruction: End-to-End Athlete Tracking and Identification on a Minimap

Transferable Knowledge-Based Multi-Granularity Aggregation Network for Temporal Action Localization: Submission to ActivityNet Challenge 2021

Towards Universal Soccer Video Understanding

YH Technologies at ActivityNet Challenge 2018

MSR Asia MSM at ActivityNet Challenge 2017: Trimmed Action Recognition, Temporal Action Proposals and Dense-Captioning Events in Videos

Two-Stream Consensus Network: Submission to HACS Challenge 2021 Weakly-Supervised Learning Track