Research map / Image generation / Text-to-image diffusion
Text-guided Controllable Diffusion for Realistic Camouflage Images Generation
Yuhang Qian, Haiyan Chen, Wentong Li, Ningzhong Liu, Jie Qin
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 Cross Normalization (with FIRM) (ablation) | CT-CIG · COD10K camouflage generation | FID: 33.99 -> 30.59 | Improved |
| add FIRM (ablation) | CT-CIG · COD10K camouflage generation | FID: 31.66 -> 30.59 | Improved |
| add FIRM and CN together (ablation) | CT-CIG · COD10K camouflage generation | KID: 0.0079 -> 0.0085 | Hurt |
| add FIRM and CN together (ablation) | CT-CIG · COD10K camouflage generation | FID: 32.37 -> 30.59 | Improved |
| train on detailed prompts instead of simple prompts (ablation) | CT-CIG · COD10K camouflage generation | FID: 54.92 -> 30.59 | Improved |
| forbid outline mentions in generated prompts (silent outline) (ablation) | CT-CIG · COD10K camouflage generation | FID: 39.79 -> 30.59 | Improved |
| include object outline in VLM prompt generation (silent outline vs no outline) (ablation) | CT-CIG · COD10K camouflage generation | CLIPScore: 0.3247 -> 0.3242 | No clear effect |
| use detailed rather than simple prompts for VLM captioning (ablation) | CRDM · CLIPScore across VLMs | CLIPScore: LLaVA-13B 0.2986 -> 0.2969; Qwen2.5-VL 0.3183 -> 0.3242 | Mixed |
| use CT-CIG instead of SOO | camouflage generation · LAKE-RED Camouflaged Objects | KID: 0.0056 -> 0.0085 | Hurt |
| use CT-CIG instead of LAKERED | camouflage generation · LAKE-RED Overall | FID: 64.27 -> 52.88 | Improved |
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.
- ControlNet — ControlNet++: Improving Conditional Controls with Efficient Consistency Feedback (ECCV 2024) · better on 5 of 5 shared results
- LAKERED — LAKE-RED: Camouflaged Images Generation by Latent Background Knowledge Retrieval-Augmented Diffusion (CVPR 2024) · better on 4 of 4 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
- MMCSBench: A Fine-Grained Benchmark for Large Vision-Language Models in Camouflage Scenes (NeurIPS 2025 · similarity 0.87)
- Deep Camouflage Images (AAAI 2020 · similarity 0.86)
- MM-CamObj: A Comprehensive Multimodal Dataset for Camouflaged Object Scenarios (AAAI 2025 · similarity 0.86)
- Camouflage Anything: Learning to Hide using Controlled Out-painting and Representation Engineering (CVPR 2025 · similarity 0.86)
- LAKE-RED: Camouflaged Images Generation by Latent Background Knowledge Retrieval-Augmented Diffusion (CVPR 2024 · similarity 0.85)
- CamoDiffusion: Camouflaged Object Detection via Conditional Diffusion Models (AAAI 2024 · 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.
