TY - JOUR
T1 - Weakly Supervised Semantic Segmentation via Alternate Self-Dual Teaching
AU - Zhang, Dingwen
AU - Li, Hao
AU - Zeng, Wenyuan
AU - Fang, Chaowei
AU - Cheng, Lechao
AU - Cheng, Ming Ming
AU - Han, Junwei
N1 - Publisher Copyright:
© 1992-2012 IEEE.
PY - 2025
Y1 - 2025
N2 - Weakly supervised semantic segmentation (WSSS) is a challenging yet important research field in vision community. In WSSS, the key problem is to generate high-quality pseudo segmentation masks (PSMs). Existing approaches mainly depend on the discriminative object part to generate PSMs, which would inevitably miss object parts or involve surrounding image background, as the learning process is unaware of the full object structure. In fact, both the discriminative object part and the full object structure are critical for deriving of high-quality PSMs. To fully explore these two information cues, we build a novel end-to-end learning framework, alternate self-dual teaching (ASDT), based on a dual-teacher single-student network architecture. The information interaction among different network branches is formulated in the form of knowledge distillation (KD). Unlike the conventional KD, the knowledge of the two teacher models would inevitably be noisy under weak supervision. Inspired by the Pulse Width (PW) modulation, we introduce a PW wave-like selection signal to alleviate the influence of the imperfect knowledge from either teacher model on the KD process. Comprehensive experiments on the PASCAL VOC 2012 and COCO-Stuff 10K demonstrate the effectiveness of the proposed ASDT framework, and new state-of-the-art results are achieved.
AB - Weakly supervised semantic segmentation (WSSS) is a challenging yet important research field in vision community. In WSSS, the key problem is to generate high-quality pseudo segmentation masks (PSMs). Existing approaches mainly depend on the discriminative object part to generate PSMs, which would inevitably miss object parts or involve surrounding image background, as the learning process is unaware of the full object structure. In fact, both the discriminative object part and the full object structure are critical for deriving of high-quality PSMs. To fully explore these two information cues, we build a novel end-to-end learning framework, alternate self-dual teaching (ASDT), based on a dual-teacher single-student network architecture. The information interaction among different network branches is formulated in the form of knowledge distillation (KD). Unlike the conventional KD, the knowledge of the two teacher models would inevitably be noisy under weak supervision. Inspired by the Pulse Width (PW) modulation, we introduce a PW wave-like selection signal to alleviate the influence of the imperfect knowledge from either teacher model on the KD process. Comprehensive experiments on the PASCAL VOC 2012 and COCO-Stuff 10K demonstrate the effectiveness of the proposed ASDT framework, and new state-of-the-art results are achieved.
KW - Weakly supervised learning
KW - dual teaching
KW - knowledge distillation
KW - semantic segmentation
UR - http://www.scopus.com/inward/record.url?scp=85182371189&partnerID=8YFLogxK
U2 - 10.1109/TIP.2023.3343112
DO - 10.1109/TIP.2023.3343112
M3 - 文章
AN - SCOPUS:85182371189
SN - 1057-7149
VL - 34
SP - 3086
EP - 3095
JO - IEEE Transactions on Image Processing
JF - IEEE Transactions on Image Processing
ER -