Research map / Detection and segmentation / Weakly supervised detection and segmentation
From Easy to Hard: Progressive Active Learning Framework for Infrared Small Target Detection with Single Point Supervision
Chuang Yu, Jinmiao Zhao, Yunpeng Liu, Sicheng Zhao, Yimian Dai, Xiangyu Yue
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| add model pre-start with EPG easy samples (ablation) | PAL (MSDA-Net) · SIRST3-Test (coarse point) | IoU (%): 1.23 -> 56.24 | Improved |
| add coarse outer updates (COU) to pre-start (ablation) | PAL (MSDA-Net) · SIRST3-Test (coarse point) | IoU (%): 56.24 -> 61.64 | Improved |
| add fine inner updates (FIU) to pre-start (ablation) | PAL (MSDA-Net) · SIRST3-Test (coarse point) | IoU (%): 56.24 -> 61.66 | Improved |
| combine COU and FIU (fine dual-update) (ablation) | PAL (MSDA-Net) · SIRST3-Test (coarse point) | IoU (%): 61.66 -> 69.38 | Improved |
| add decay factor lambda=0.97 (ablation) | PAL (MSDA-Net) · SIRST3-Test (coarse point) | IoU (%): 67.90 -> 69.38 | Improved |
| use EEDM loss instead of Dice loss (ablation) | PAL (MSDA-Net) · SIRST3-Test (coarse point) | IoU (%): 64.38 -> 69.38 | Improved |
| use EEDM loss instead of BCE loss (ablation) | PAL (MSDA-Net) · SIRST3-Test (coarse point) | IoU (%): 69.26 -> 69.38 | No clear effect |
| use EEDM loss instead of Focal loss (ablation) | PAL (MSDA-Net) · SIRST3-Test (coarse point) | IoU (%): 68.89 -> 69.38 | Improved |
| select only easy samples initially (MP) vs all samples (MP-1) (ablation) | PAL (MSDA-Net) · SIRST3-Test (coarse point) | IoU (%): 32.86 -> 56.24 | Improved |
| gradually add hard samples via COU on top of easy-only training with FIU (ablation) | PAL (MSDA-Net) · SIRST3-Test (coarse point) | IoU (%): 61.66 -> 69.38 | Improved |
| update period 5 instead of 1 (ablation) | PAL (MSDA-Net) · SIRST3-Test (coarse point) | IoU (%): 66.35 -> 69.38 | Improved |
| learning rate 1e-3 instead of 5e-5 (ablation) | PAL (MSDA-Net) · SIRST3-Test (coarse point) | IoU (%): 59.68 -> 69.38 | Improved |
Closest papers by meaning
- Semi-supervised Infrared Small Target Detection with Thermodynamic-Inspired Uneven Perturbation and Confidence Adaptation (AAAI 2025 · similarity 0.90)
- Rethinking Infrared Small Target Detection: A Foundation Driven Efficient Paradigm (NeurIPS 2026 · similarity 0.89)
- Rethinking Evaluation of Infrared Small Target Detection (NeurIPS 2025 · similarity 0.86)
- Pinwheel-shaped Convolution and Scale-based Dynamic Loss for Infrared Small Target Detection (AAAI 2025 · similarity 0.85)
- IRPruneDet: Efficient Infrared Small Target Detection via Wavelet Structure-Regularized Soft Channel Pruning (AAAI 2024 · similarity 0.85)
- Motion Prior Knowledge Learning with Homogeneous Language Descriptions for Moving Infrared Small Target Detection (AAAI 2025 · similarity 0.85)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
