TY - GEN
T1 - Harmonizing Classification and Localization in Small Object Detection
AU - Chai, Enhui
AU - Chen, Li
AU - Wei, Liu
AU - Cui, Tianxiang
N1 - Publisher Copyright:
© The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2026.
PY - 2026
Y1 - 2026
N2 - In the field of Small Object Detection (SOD), accurate classification and localization are crucial for detection performance. However, the inherent imbalance between classification and localization tasks can generate conflicting priorities, leading to suboptimal task coordination for small object detection. This imbalance is mainly caused by the different attention regions and the gradient competition between the two tasks during joint training. In this paper, we propose a Dual-Task Harmonization Framework (DTHF): First, we introduce a Feature Fusion-based Data Augmentation strategy (FF-DA), which amplifies boundary-aware patterns for localization while preserving critical semantic regions for classification, thereby aligning their region-of-interest priorities. Second, we design a Gradient Equilibrium Module (GEM) that dynamically balances tasks by altering the gradients, preventing one task from overwhelming the other during optimization. Experiments on the MS COCO and VisDrone datasets demonstrate that our method, compared to the baseline model, the experimental metrics of our method mAP in the VisDrone data set are improved by 2.0+%. Ablation studies validate that both FF-DA and GEM contribute synergistically, offering a unified solution to task imbalance in small object detection.
AB - In the field of Small Object Detection (SOD), accurate classification and localization are crucial for detection performance. However, the inherent imbalance between classification and localization tasks can generate conflicting priorities, leading to suboptimal task coordination for small object detection. This imbalance is mainly caused by the different attention regions and the gradient competition between the two tasks during joint training. In this paper, we propose a Dual-Task Harmonization Framework (DTHF): First, we introduce a Feature Fusion-based Data Augmentation strategy (FF-DA), which amplifies boundary-aware patterns for localization while preserving critical semantic regions for classification, thereby aligning their region-of-interest priorities. Second, we design a Gradient Equilibrium Module (GEM) that dynamically balances tasks by altering the gradients, preventing one task from overwhelming the other during optimization. Experiments on the MS COCO and VisDrone datasets demonstrate that our method, compared to the baseline model, the experimental metrics of our method mAP in the VisDrone data set are improved by 2.0+%. Ablation studies validate that both FF-DA and GEM contribute synergistically, offering a unified solution to task imbalance in small object detection.
KW - Data augmentation
KW - Gradient equilibrium module
KW - Region-of-interest
KW - Small object detection
KW - Task imbalance
UR - https://www.scopus.com/pages/publications/105022903836
U2 - 10.1007/978-981-95-4097-6_29
DO - 10.1007/978-981-95-4097-6_29
M3 - Conference contribution
AN - SCOPUS:105022903836
SN - 9789819540969
T3 - Communications in Computer and Information Science
SP - 426
EP - 440
BT - Neural Information Processing - 32nd International Conference, ICONIP 2025, Proceedings
A2 - Taniguchi, Tadahiro
A2 - Leung, Chi Sing Andrew
A2 - Kozuno, Tadashi
A2 - Yoshimoto, Junichiro
A2 - Mahmud, Mufti
A2 - Doborjeh, Maryam
A2 - Doya, Kenji
PB - Springer Science and Business Media Deutschland GmbH
T2 - 32nd International Conference on Neural Information Processing, ICONIP 2025
Y2 - 20 November 2025 through 24 November 2025
ER -