HC-STVG1

Human-centric Spatio-Temporal Video Grounding

Introduced 2020-11-10

The newly proposed HC-STVG task aims to localize the target person spatio-temporally in an untrimmed video. For this task, we collect a new benchmark dataset, which has spatio temporal annotations related to the target persons in complex multi-person scenes, together with full interaction and rich action information.

Benchmarks

Spatio-Temporal Video Grounding/m_vIoU Spatio-Temporal Video Grounding/vIoU@0.3 Spatio-Temporal Video Grounding/vIoU@0.5