TY - GEN
T1 - ADAPTIVE INTRA-GROUP AGGREGATION FOR CO-SALIENCY DETECTION
AU - Ren, Guangyu
AU - Dai, Tianhong
AU - Stathaki, Tania
N1 - Publisher Copyright:
© 2022 IEEE
PY - 2022
Y1 - 2022
N2 - Co-salient object detection (CoSOD) together with the rapid development of deep learning has led to substantial progress in recent years. However, the feature aggregation between group feature representation and individual feature representation is still a challenging issue. In this work, we propose a novel adaptive intra-group aggregation (AIGA) method, which provides a new perspective to investigate the interaction relationship between group and single-image features and aggregate these features in an adaptive way. A novel scale-aware loss is proposed to help the model capture the scale prior of different groups and discriminatively process groups during the training phase. Extensive experiments demonstrate that the proposed method can effectively improve the performance without increasing extra parameters and achieve better accuracy on three prevalent benchmarks.
AB - Co-salient object detection (CoSOD) together with the rapid development of deep learning has led to substantial progress in recent years. However, the feature aggregation between group feature representation and individual feature representation is still a challenging issue. In this work, we propose a novel adaptive intra-group aggregation (AIGA) method, which provides a new perspective to investigate the interaction relationship between group and single-image features and aggregate these features in an adaptive way. A novel scale-aware loss is proposed to help the model capture the scale prior of different groups and discriminatively process groups during the training phase. Extensive experiments demonstrate that the proposed method can effectively improve the performance without increasing extra parameters and achieve better accuracy on three prevalent benchmarks.
KW - Co-salient object detection
KW - determinantal point processes
UR - http://www.scopus.com/inward/record.url?scp=85131258548&partnerID=8YFLogxK
U2 - 10.1109/ICASSP43922.2022.9746218
DO - 10.1109/ICASSP43922.2022.9746218
M3 - Conference Proceeding
AN - SCOPUS:85131258548
T3 - ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings
SP - 2520
EP - 2524
BT - 2022 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2022 - Proceedings
PB - Institute of Electrical and Electronics Engineers Inc.
T2 - 47th IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2022
Y2 - 23 May 2022 through 27 May 2022
ER -