Tiny Video Networks

摘要

Abstract Automatic video understanding is becoming more important for applications where real‐time performance is crucial and compute is limited: for example, automated video tagging, robot perception, activity recognition for mobile devices. Yet, accurate solutions so far have been computationally intensive. We propose efficient models for videos—Tiny Video Networks—which are video architectures, automatically designed to comply with fast runtimes and, at the same time are effective at video recognition tasks. The TVNs run at faster‐than‐real‐time speeds and demonstrate strong performance across several video benchmarks. These models not only provide new tools for real‐time video applications, but also enable fast research and development in video understanding. Code and models are available.