DeCo: Decomposition and Reconstruction for Compositional Temporal Grounding via Coarse-to-Fine Contrastive Ranking

Understanding dense action in videos is a fundamental challenge towards the generalization of vision models. Several works show that compositionality is key to achieving generalization by combining known primitive elements, especially for handling novel composited structures. Compositional temporal...

Full description

Saved in:
Bibliographic Details
Main Authors: Yang, Lijin, Kong, Quan, Yang, Hsuan-Kung, Kehl, Wadim, Sato, Yoichi, Kobori, Norimasa
Format: Conference Proceeding
Language:eng ; jpn
Subjects:
and reasoning
Cognition
Computer vision
Grounding
language
Pattern recognition
Semantics
Task analysis
Videos
Vision
Online Access:Request full text
Tags: Add Tag
No Tags, Be the first to tag this record!