10 teams for
Track #1: Spatiotemporal Action Localization (Chaotic World dataset)
-
8 teams for
Track #2: Behavioral Graph Analysis (Chaotic World dataset)
+
8 teams for
Track #2: (Scene Graph Generation) (Chaotic World dataset)
12 teams for
Track #3: Spatiotemporal Event Grounding (Chaotic World dataset)
10 teams for
Track #4: Sound Source Localization (Chaotic World dataset)
13 teams for
Track #5: Video Grounding (Animal Kingdom dataset)
@@ -148,7 +148,7 @@
Challenge Details
Chaotic World (https://github.com/sutdcv/Chaotic-World)
This challenging multi-modal (video, audio, and text) video dataset that focuses on chaotic situations around the world comprises complex and dynamic scenes with severe occlusions and contains over 200,000 annotated instances for various tasks such as spatiotemporal action localization (i.e., spatially and temporally locate the action), behavioral graph analysis (i.e., analyze interactions between people), spatiotemporal event grounding (i.e., identifying relevant segments in long videos and localizing people, scene, and behavior-of-interest), and sound source localization (i.e., spatially locate the source of sound). This will promote deeper research into more robust models that capitalize various modalities and handle such complex human behaviors / interactions in dynamic and complex environments.
Track #1: Spatiotemporal Action Localization
-
Track #2: Behavioral Graph Analysis
+
Track #2: Behavioral Graph Analysis (Scene Graph Generation)
Track #3: Spatiotemporal Event Grounding
Track #4: Sound Source Localization
