TY - GEN
T1 - DAP
T2 - 2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition, CVPR 2021
AU - Zhong, Yuanyi
AU - Wang, Jianfeng
AU - Wang, Lijuan
AU - Peng, Jian
AU - Wang, Yu Xiong
AU - Zhang, Lei
N1 - Publisher Copyright:
© 2021 IEEE
PY - 2021
Y1 - 2021
N2 - This paper presents a detection-aware pre-training (DAP) approach, which leverages only weakly-labeled classification-style datasets (e.g., ImageNet) for pretraining, but is specifically tailored to benefit object detection tasks. In contrast to the widely used image classification-based pre-training (e.g., on ImageNet), which does not include any location-related training tasks, we transform a classification dataset into a detection dataset through a weakly supervised object localization method based on Class Activation Maps to directly pre-train a detector, making the pre-trained model location-aware and capable of predicting bounding boxes. We show that DAP can outperform the traditional classification pre-training in terms of both sample efficiency and convergence speed in downstream detection tasks including VOC and COCO. In particular, DAP boosts the detection accuracy by a large margin when the number of examples in the downstream task is small.
AB - This paper presents a detection-aware pre-training (DAP) approach, which leverages only weakly-labeled classification-style datasets (e.g., ImageNet) for pretraining, but is specifically tailored to benefit object detection tasks. In contrast to the widely used image classification-based pre-training (e.g., on ImageNet), which does not include any location-related training tasks, we transform a classification dataset into a detection dataset through a weakly supervised object localization method based on Class Activation Maps to directly pre-train a detector, making the pre-trained model location-aware and capable of predicting bounding boxes. We show that DAP can outperform the traditional classification pre-training in terms of both sample efficiency and convergence speed in downstream detection tasks including VOC and COCO. In particular, DAP boosts the detection accuracy by a large margin when the number of examples in the downstream task is small.
UR - http://www.scopus.com/inward/record.url?scp=85123163893&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85123163893&partnerID=8YFLogxK
U2 - 10.1109/CVPR46437.2021.00451
DO - 10.1109/CVPR46437.2021.00451
M3 - Conference contribution
AN - SCOPUS:85123163893
T3 - Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition
SP - 4535
EP - 4544
BT - Proceedings - 2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition, CVPR 2021
PB - IEEE Computer Society
Y2 - 19 June 2021 through 25 June 2021
ER -