Weakly supervised learning of interactions between humans and objects

Alessandro Prest 1, 2 Cordelia Schmid 1 Vittorio Ferrari 2
1 LEAR - Learning and recognition in vision
Inria Grenoble - Rhône-Alpes, LJK - Laboratoire Jean Kuntzmann, INPG - Institut National Polytechnique de Grenoble
Abstract : We introduce a weakly supervised approach for learning human actions modeled as interactions between humans and objects. Our approach is human-centric: we first localize a human in the image and then determine the object relevant for the action and its spatial relation with the human. The model is learned automatically from a set of still images annotated (only) with the action label. Our approach relies on a human detector to initialize the model learning. For robustness to various degrees of visibility, we build a detector that learns to combine a set of existing part detectors. Starting from humans detected in a set of images depicting the action, our approach determines the action object and its spatial relation to the human. Its final output is a probabilistic model of the human-object interaction, i.e. the spatial relation between the human and the object.
Liste complète des métadonnées

Cited literature [35 references]  Display  Hide  Download


https://hal.inria.fr/inria-00516477
Contributor : Alessandro Prest <>
Submitted on : Thursday, September 9, 2010 - 5:28:21 PM
Last modification on : Tuesday, March 5, 2019 - 9:30:11 AM
Document(s) archivé(s) le : Tuesday, October 23, 2012 - 3:50:20 PM

Files

RT-0391.pdf
Files produced by the author(s)

Identifiers

  • HAL Id : inria-00516477, version 1

Collections

Citation

Alessandro Prest, Cordelia Schmid, Vittorio Ferrari. Weakly supervised learning of interactions between humans and objects. [Technical Report] RT-391, INRIA. 2010. ⟨inria-00516477⟩

Share

Metrics

Record views

604

Files downloads

588