Paper in TPAMI: Text-driven video acceleration

Our work “Text-driven video acceleration: A weakly-supervised reinforcement learning method” was accepted at the IEEE Transactions on Pattern Analysis and Machine Intelligence. The paper presents a novel method to automatically accelerate instructional videos based on the corresponding textual instructions. Check our video below:

The paper is available here. The source code of our algorithm is also available on Github at https://github.com/verlab/TextDrivenVideoAcceleration_TPAMI_2022.