other · Underline Science Inc. · 2021

StarNet: Towards Weakly Supervised Few-Shot Object Detection

Association for Artificial Intelligence 2021, Alfassy, Amit, Bronstein, Alex, Doveh, Sivan, Feris, Rogerio, Giryes, Raja, Harary, Sivan, Karlinsky, Leonid, Lichtenstein, Moshe, Sattigeri, Prasanna, Schwartz, Eli, Shtok, Joseph

Cited by 0Open access

Abstract

Few-shot detection and classification have advanced significantly in recent years. Yet, detection approaches require strong annotation (bounding boxes) both for pre-training and for adaptation to novel classes, and classification approaches rarely provide localization of objects in the scene. In this paper, we introduce StarNet - a few-shot model featuring an end-to-end differentiable non-parametric star-model detection and classification head. Through this head, the backbone is meta-trained using only image-level labels to produce good features for jointly localizing and classifying previously unseen categories of few-shot test tasks using a star-model that geometrically matches between the query and support images (to find corresponding object instances). Being a few-shot detector, StarNet does not require any bounding box annotations, neither during pre-training nor for novel classes adaptation. It can thus be applied to the previously unexplored and challenging task of Weakly Supervised Few-Shot Object Detection (WS-FSOD), where it attains significant improvements over the baselines. In addition, StarNet shows significant gains on few-shot classification benchmarks that are less cropped around the objects (where object localization is key).

Cite this paper

APA 7

2021, A. F. A. I., Alfassy, A., Bronstein, A., Doveh, S., Feris, R., Giryes, R., Harary, S., Karlinsky, L., Lichtenstein, M., Sattigeri, P., Schwartz, E., & Shtok, J. (2021). StarNet: Towards Weakly Supervised Few-Shot Object Detection. Underline Science Inc.. https://doi.org/10.48448/kt7a-k752

MLA 9

2021, Association for Artificial Intelligence, et al. “StarNet: Towards Weakly Supervised Few-Shot Object Detection.” Underline Science Inc., 2021, https://doi.org/10.48448/kt7a-k752.

Chicago

2021, Association for Artificial Intelligence, Amit Alfassy, Alex Bronstein, Sivan Doveh, Rogerio Feris, Raja Giryes, Sivan Harary, et al. 2021. “StarNet: Towards Weakly Supervised Few-Shot Object Detection.” Underline Science Inc.. https://doi.org/10.48448/kt7a-k752.

Harvard

2021, A.F.A.I. et al. (2021) ‘StarNet: Towards Weakly Supervised Few-Shot Object Detection’, Underline Science Inc.. Available at: https://doi.org/10.48448/kt7a-k752.

BibTeX

@misc{20212021starnet,
  title = {{StarNet: Towards Weakly Supervised Few-Shot Object Detection}},
  author = {2021, Association for Artificial Intelligence and Alfassy, Amit and Bronstein, Alex and Doveh, Sivan and Feris, Rogerio and Giryes, Raja and Harary, Sivan and Karlinsky, Leonid and Lichtenstein, Moshe and Sattigeri, Prasanna and Schwartz, Eli and Shtok, Joseph},
  journal = {Underline Science Inc.},
  year = {2021},
  doi = {10.48448/kt7a-k752}
}

Made from the index record. Check names, capitalisation and page numbers against the paper. Need another style? AI for Research has over 10,000.