Research map / Detection and segmentation / Open-set and few-shot detection
Simulating Distribution Dynamics: Liquid Temporal Feature Evolution for Single-Domain Generalized Object Detection
Zihao Zhang, Yang Li, Aming Wu, Yahong Han
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 progressive temporal feature evolution (PTFE) alone (ablation) | Faster R-CNN · Diverse Weather: Night Sunny | mAP (%): 35.4 -> 36.8 | Improved |
| add temporal dependency modeling and liquid parameter evolution on top of PTFE (ablation) | LTFE · Diverse Weather: Night Sunny | mAP (%): 36.8 -> 42.6 | Improved |
| add temporal feature alignment module (TFAM) on top of PTFE (ablation) | LTFE · Diverse Weather: Night Sunny | mAP (%): 36.8 -> 40.2 | Improved |
| add TFAM to PTFE + TDM + LPE (full model) (ablation) | LTFE · Diverse Weather: Night Rainy | mAP (%): 22.7 -> 24.3 | Improved |
| increase training evolution steps T up to 8 (ablation) | PTFE · Day Foggy and Dusk Rainy | mAP (%) | Improved |
| use larger T at inference (ablation) | LTFE · Diverse Weather | mAP (%) | Hurt |
| swap ResNet-101 for Swin-T backbone | LTFE · Diverse Weather: Night Sunny | mAP (%): 43.1 -> 53.1 | Improved |
| compare LTFE with test-time adaptation methods | LTFE vs ECPG · Diverse Weather: Night Sunny | mAP (%): 35.3 -> 43.1 | Improved |
| LTFE vs SECOT with ResNet-101 on Real-to-Art Clipart | LTFE · Clipart1k | mAP (%): 40.2 -> 42.1 | Mixed |
| LTFE vs prior methods on Diverse Weather (ResNet-101) | LTFE · Diverse Weather: Night Sunny / Night Rainy | mAP (%): Night Sunny 45.0 (G-NAS) vs 43.1; Night Rainy 24.5 (SECOT) vs 24.3 | Mixed |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 2 of 2 on most shared results.
- Faster R-CNN — Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks (TPAMI 2016) · better on 2 of 2 shared results
- G-NAS — G-NAS: Generalizable Neural Architecture Search for Single Domain Generalization Object Detection (AAAI 2024) · better on 1 of 1 shared result
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
- Style Evolving along Chain-of-Thought for Unknown-Domain Object Detection (CVPR 2025 · similarity 0.89)
- PhysAug: A Physical-guided and Frequency-based Data Augmentation for Single-Domain Generalized Object Detection (AAAI 2025 · similarity 0.88)
- Generalized Diffusion Detector: Mining Robust Features from Diffusion Models for Domain-Generalized Detection (CVPR 2025 · similarity 0.88)
- Towards Single-Source Domain Generalized Object Detection via Causal Visual Prompts (NeurIPS 2025 · similarity 0.87)
- Benchmarking Object Detectors under Real-World Distribution Shifts in Satellite Imagery (CVPR 2025 · similarity 0.87)
- Improving Single Domain-Generalized Object Detection: A Focus on Diversification and Alignment (CVPR 2024 · similarity 0.86)
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.
