Concept Generalization in Visual Representation Learning - Inria - Institut national de recherche en sciences et technologies du numérique Access content directly
Conference Papers Year : 2021

Concept Generalization in Visual Representation Learning


Measuring concept generalization, i.e., the extent to which models trained on a set of (seen) visual concepts can be leveraged to recognize a new set of (unseen) concepts, is a popular way of evaluating visual representations, especially in a self-supervised learning framework. Nonetheless, the choice of unseen concepts for such an evaluation is usually made arbitrarily, and independently from the seen concepts used to train representations, thus ignoring any semantic relationships between the two. In this paper, we argue that the semantic relationships between seen and unseen concepts affect generalization performance and propose ImageNet-CoG a novel benchmark on the ImageNet-21K (IN-21K) dataset that enables measuring concept generalization in a principled way. Our benchmark leverages expert knowledge that comes from WordNet in order to define a sequence of unseen IN-21K concept sets that are semantically more and more distant from the ImageNet-1K (IN-1K) subset, a ubiquitous training set. This allows us to benchmark visual representations learned on IN-1K out-of-the box. We conduct a large-scale study encompassing 31 convolution and transformer-based models and show how different architectures, levels of supervision, regularization techniques and use of web data impact theconcept generalization performance.
Fichier principal
Vignette du fichier
01015.pdf (982.05 Ko) Télécharger le fichier
Origin Files produced by the author(s)

Dates and versions

hal-03110632 , version 1 (14-01-2021)
hal-03110632 , version 2 (07-09-2021)



Mert Bulent Sariyildiz, Yannis Kalantidis, Diane Larlus, Karteek Alahari. Concept Generalization in Visual Representation Learning. ICCV 2021 - International Conference on Computer Vision, Oct 2021, Virtual, Canada. ⟨hal-03110632v2⟩
195 View
310 Download



Gmail Mastodon Facebook X LinkedIn More