Joint learning of object and action detectors - Inria - Institut national de recherche en sciences et technologies du numérique Access content directly
Conference Papers Year : 2017

Joint learning of object and action detectors


While most existing approaches for detection in videos focus on objects or human actions separately, we aim at jointly detecting objects performing actions, such as cat eating or dog jumping. We introduce an end-to-end multitask objective that jointly learns object-action relationships. We compare it with different training objectives, validate its effectiveness for detecting objects-actions in videos, and show that both tasks of object and action detection benefit from this joint learning. Moreover, the proposed architecture can be used for zero-shot learning of actions: our multitask objective leverages the commonalities of an action performed by different objects, e.g. dog and cat jumping , enabling to detect actions of an object without training with these object-actions pairs. In experiments on the A2D dataset [50], we obtain state-of-the-art results on segmentation of object-action pairs. We finally apply our multitask architecture to detect visual relationships between objects in images of the VRD dataset [24].
Fichier principal
Vignette du fichier
main.pdf (5.68 Mo) Télécharger le fichier
Vignette du fichier
net.jpg (118.42 Ko) Télécharger le fichier
Origin Files produced by the author(s)
Format Figure, Image
Origin Files produced by the author(s)

Dates and versions

hal-01575804 , version 1 (21-08-2017)



Vicky Kalogeiton, Philippe Weinzaepfel, Vittorio Ferrari, Cordelia Schmid. Joint learning of object and action detectors. ICCV - IEEE International Conference on Computer Vision, Oct 2017, Venice, Italy. pp.2001-2010, ⟨10.1109/ICCV.2017.219⟩. ⟨hal-01575804⟩
969 View
3124 Download



Gmail Mastodon Facebook X LinkedIn More