TRIM is a self-supervised video summarization framework that uses Markov-driven losses to achieve state-of-the-art unsupervised results on SUMME and TVSUM while rivaling top supervised models.
Masked autoencoder for unsupervised video summarization
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
fields
cs.CV 2verdicts
UNVERDICTED 2representative citing papers
TRIMMER proposes a self-supervised RL method for video summarization that uses entropy-based rewards to capture temporal dynamics and semantic diversity, claiming SOTA results among unsupervised approaches.
citing papers explorer
-
TRIM: A Self-Supervised Video Summarization Framework Maximizing Temporal Relative Information and Representativeness
TRIM is a self-supervised video summarization framework that uses Markov-driven losses to achieve state-of-the-art unsupervised results on SUMME and TVSUM while rivaling top supervised models.
-
TRIMMER: A New Paradigm for Video Summarization through Self-Supervised Reinforcement Learning
TRIMMER proposes a self-supervised RL method for video summarization that uses entropy-based rewards to capture temporal dynamics and semantic diversity, claiming SOTA results among unsupervised approaches.