Argus++: Robust Real-time Activity Detection for Unconstrained Video Streams with Overlapping Cube Proposals

TL;DR

Argus++ is a real-time activity detection system for unconstrained video streams that uses overlapping spatio-temporal cubes to improve coverage and robustness, outperforming existing methods on multiple benchmarks.

Contribution

It introduces overlapping cube proposals and an optimized system design for real-time, robust activity detection in untrimmed, large field-of-view videos.

Findings

Superior performance on multiple activity detection benchmarks

Effective coverage of multi-scale, multi-instance activities

Real-time processing on consumer hardware

Abstract

Activity detection is one of the attractive computer vision tasks to exploit the video streams captured by widely installed cameras. Although achieving impressive performance, conventional activity detection algorithms are usually designed under certain constraints, such as using trimmed and/or object-centered video clips as inputs. Therefore, they failed to deal with the multi-scale multi-instance cases in real-world unconstrained video streams, which are untrimmed and have large field-of-views. Real-time requirements for streaming analysis also mark brute force expansion of them unfeasible. To overcome these issues, we propose Argus++, a robust real-time activity detection system for analyzing unconstrained video streams. The design of Argus++ introduces overlapping spatio-temporal cubes as an intermediate concept of activity proposals to ensure coverage and completeness of activity detection through over-sampling. The overall system is optimized for real-time processing on standalone consumer-level hardware. Extensive experiments on different surveillance and driving scenarios demonstrated its superior performance in a series of activity detection benchmarks, including CVPR ActivityNet ActEV 2021, NIST ActEV SDL UF/KF, TRECVID ActEV 2020/2021, and ICCV ROAD 2021.