Research map / Video / Video segmentation and tracking
Sparsely Timing the Change: A Spiking Temporal Framework for Remote Sensing Interpretation
Shilong Li, Xiurui Xie, Qiugang Zhan, Luochao Wang, Yong Deng, Guisong Liu
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 SpikeAdapter (GSI-P + SNN + STSpikeFuse) to a ViT-SAM Large backbone | change detection backbone · CLCD | F1 (%): 78.03 -> 82.96 (vs Change3D) | Improved |
| add SpikeAdapter | change detection backbone · SYSU-CD | F1 (%): 82.66 -> 84.11 (vs CD-Lamba) | Improved |
| add GSI-P with STSpikeFuse to TTP baseline (ablation) | TTP · LEVIR-CD | F1 (%): 92.10 -> 93.08 | Improved |
| use linear delay mapping instead of inverse or learned MLP (ablation) | GSI-P · LEVIR-CD | F1 (%): 92.84 / 92.88 -> 93.08 | Improved |
| add triangular temporal smoothing kernel (ablation) | GSI-P · LEVIR-CD | F1 (%): 92.65 -> 93.08 | Improved |
| use triangular kernel instead of Gaussian kernel (ablation) | GSI-P · LEVIR-CD | F1 (%): 92.85 -> 93.08 | Improved |
| use STSpikeFuse instead of simple addition of SNN features (ablation) | SpikeAdapter · LEVIR-CD | F1 (%): 92.40 -> 93.08 | Improved |
| use STSpikeFuse instead of concatenation (ablation) | SpikeAdapter · LEVIR-CD | F1 (%): 92.87 -> 93.08 | Improved |
| fuse attention output with backbone feature rather than using it directly (ablation) | STSpikeFuse · LEVIR-CD | F1 (%): 75.67 -> 92.8 | Improved |
| use learnable gate instead of addition or concatenation for final fusion (ablation) | STSpikeFuse · LEVIR-CD | F1 (%): 92.82 (Add) / 92.95 (Concat) vs 92.8 (Gate) | No clear effect |
| raise input resolution from 256 to 512 (ablation) | SpikeAdapter · LEVIR-CD | F1 (%): 92.96 -> 93.08 | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- Change3D — Change3D: Revisiting Change Detection and Captioning from A Video Modeling Perspective (CVPR 2025) · better on 3 of 3 shared results
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- Spiking Meets Attention: Efficient Remote Sensing Image Super-Resolution with Attention Spiking Neural Networks (NeurIPS 2025 · similarity 0.88)
- FM-ChangeNet: Learning Change through Pathwise Feature Transport (NeurIPS 2026 · similarity 0.84)
- Efficient Representation Learning of Satellite Image Time Series and Their Fusion for Spatiotemporal Applications (AAAI 2024 · similarity 0.83)
- Deep Random Features for Scalable Interpolation of Spatiotemporal Data (ICLR 2025 · similarity 0.83)
- SkySense: A Multi-Modal Remote Sensing Foundation Model Towards Universal Interpretation for Earth Observation Imagery (CVPR 2024 · similarity 0.83)
- STAA-SNN: Spatial-Temporal Attention Aggregator for Spiking Neural Networks (CVPR 2025 · similarity 0.83)
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.
