TY - GEN
T1 - StarNet
T2 - 35th AAAI Conference on Artificial Intelligence, AAAI 2021
AU - Karlinsky, Leonid
AU - Shtok, Joseph
AU - Alfassy, Amit
AU - Lichtenstein, Moshe
AU - Harary, Sivan
AU - Schwartz, Eli
AU - Doveh, Sivan
AU - Sattigeri, Prasanna
AU - Feris, Rogerio
AU - Bronstein, Alexander
AU - Giryes, Raja
N1 - Publisher Copyright:
Copyright © 2021, Association for the Advancement of Artificial Intelligence (www.aaai.org). All rights reserved
PY - 2021
Y1 - 2021
N2 - 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).
AB - 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).
UR - http://www.scopus.com/inward/record.url?scp=85121195313&partnerID=8YFLogxK
U2 - 10.1609/aaai.v35i2.16268
DO - 10.1609/aaai.v35i2.16268
M3 - ???researchoutput.researchoutputtypes.contributiontobookanthology.conference???
AN - SCOPUS:85121195313
T3 - 35th AAAI Conference on Artificial Intelligence, AAAI 2021
SP - 1743
EP - 1753
BT - 35th AAAI Conference on Artificial Intelligence, AAAI 2021
PB - Association for the Advancement of Artificial Intelligence
Y2 - 2 February 2021 through 9 February 2021
ER -