Aug 2026
Robust webly supervised fine-grained recognition via decoupled global-local fusion and geometric-semantic consensus.
A novel framework named Decoupled Global-local Consensus Learning (DGCL) is proposed that identifies and leverages noisy-labeled samples through multi-perspective analysis, enhancing both robustness and representation power.
Xin-Yi Guo, Yiwei Lu, Tao Yan
· Neural Networks · 0 citations