Grounded Human-Object Interaction Hotspots from Video (Extended Abstract)
This work addresses the challenge of embodied visual intelligence by reducing supervision needs for affordance learning, though it is incremental in leveraging existing video data.
The paper tackles the problem of learning human-object interaction hotspots from video without heavy supervision, achieving competitive performance with strongly supervised methods and enabling anticipation of interactions for novel object categories.
Learning how to interact with objects is an important step towards embodied visual intelligence, but existing techniques suffer from heavy supervision or sensing requirements. We propose an approach to learn human-object interaction "hotspots" directly from video. Rather than treat affordances as a manually supervised semantic segmentation task, our approach learns about interactions by watching videos of real human behavior and anticipating afforded actions. Given a novel image or video, our model infers a spatial hotspot map indicating how an object would be manipulated in a potential interaction, even if the object is currently at rest. Through results with both first and third person video, we show the value of grounding affordances in real human-object interactions. Not only are our weakly supervised hotspots competitive with strongly supervised affordance methods, but they can also anticipate object interaction for novel object categories. Project page: http://vision.cs.utexas.edu/projects/interaction-hotspots/