Few-Shot Learning for Road Object Detection

Anay Majee, Kshitij Agrawal, Anbumani Subramanian
AAAI Workshop on Meta-Learning and MetaDL Challenge, PMLR 140:115-126, 2021.

Abstract

Few-shot learning is a problem of high interest in the evolution of deep learning. In this work, we consider the problem of few-shot object detection (FSOD) in a real-world, class-imbalanced scenario. For our experiments, we utilize the India Driving Dataset (IDD), as it includes a class of less-occurring road objects in the image dataset and hence provides a setup suitable for few-shot learning. We evaluate both metric-learning and meta- learning based FSOD methods, in two experimental settings: (i) representative (same-domain) splits from IDD, that evaluates the ability of a model to learn in the context of road images, and (ii) object classes with less-occurring object samples, similar to the open-set setting in real-world. From our experiments, we demonstrate that the metric-learning method outperforms meta- learning on the novel classes by (i) 11.2 mAP points on the same domain, and (ii) 1.0 mAP point on the open-set. We also show that our extension of object classes in a real-world open dataset offers a rich ground for few-shot learning studies.

Cite this Paper

BibTeX
@InProceedings{pmlr-v140-majee21a, title = {Few-Shot Learning for Road Object Detection}, author = {Majee, Anay and Agrawal, Kshitij and Subramanian, Anbumani}, booktitle = {AAAI Workshop on Meta-Learning and MetaDL Challenge}, pages = {115--126}, year = {2021}, editor = {Guyon, Isabelle and van Rijn, Jan N. and Treguer, Sébastien and Vanschoren, Joaquin}, volume = {140}, series = {Proceedings of Machine Learning Research}, month = {09 Feb}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v140/majee21a/majee21a.pdf}, url = {https://proceedings.mlr.press/v140/majee21a.html}, abstract = {Few-shot learning is a problem of high interest in the evolution of deep learning. In this work, we consider the problem of few-shot object detection (FSOD) in a real-world, class-imbalanced scenario. For our experiments, we utilize the India Driving Dataset (IDD), as it includes a class of less-occurring road objects in the image dataset and hence provides a setup suitable for few-shot learning. We evaluate both metric-learning and meta- learning based FSOD methods, in two experimental settings: (i) representative (same-domain) splits from IDD, that evaluates the ability of a model to learn in the context of road images, and (ii) object classes with less-occurring object samples, similar to the open-set setting in real-world. From our experiments, we demonstrate that the metric-learning method outperforms meta- learning on the novel classes by (i) 11.2 mAP points on the same domain, and (ii) 1.0 mAP point on the open-set. We also show that our extension of object classes in a real-world open dataset offers a rich ground for few-shot learning studies.} }
Endnote
%0 Conference Paper %T Few-Shot Learning for Road Object Detection %A Anay Majee %A Kshitij Agrawal %A Anbumani Subramanian %B AAAI Workshop on Meta-Learning and MetaDL Challenge %C Proceedings of Machine Learning Research %D 2021 %E Isabelle Guyon %E Jan N. van Rijn %E Sébastien Treguer %E Joaquin Vanschoren %F pmlr-v140-majee21a %I PMLR %P 115--126 %U https://proceedings.mlr.press/v140/majee21a.html %V 140 %X Few-shot learning is a problem of high interest in the evolution of deep learning. In this work, we consider the problem of few-shot object detection (FSOD) in a real-world, class-imbalanced scenario. For our experiments, we utilize the India Driving Dataset (IDD), as it includes a class of less-occurring road objects in the image dataset and hence provides a setup suitable for few-shot learning. We evaluate both metric-learning and meta- learning based FSOD methods, in two experimental settings: (i) representative (same-domain) splits from IDD, that evaluates the ability of a model to learn in the context of road images, and (ii) object classes with less-occurring object samples, similar to the open-set setting in real-world. From our experiments, we demonstrate that the metric-learning method outperforms meta- learning on the novel classes by (i) 11.2 mAP points on the same domain, and (ii) 1.0 mAP point on the open-set. We also show that our extension of object classes in a real-world open dataset offers a rich ground for few-shot learning studies.
APA
Majee, A., Agrawal, K. & Subramanian, A.. (2021). Few-Shot Learning for Road Object Detection. AAAI Workshop on Meta-Learning and MetaDL Challenge, in Proceedings of Machine Learning Research 140:115-126 Available from https://proceedings.mlr.press/v140/majee21a.html.

Related Material