Research map / Image generation / GANs
Generating a Fusion Image: One's Identity and Another's Shape
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 |
|---|---|---|---|
| Use FusionGAN instead of CycleGAN or DiscoGAN | YouTube Pose image fusion · YouTube Pose | keypoint similarity: FusionGAN 0.52; CycleGAN 0.42; DiscoGAN 0.38 | Helped |
| Use FusionGAN instead of CycleGAN or DiscoGAN | YouTube Pose image fusion · YouTube Pose user study | reality rating: FusionGAN 3.23; CycleGAN 2.77; DiscoGAN 0.75 | Helped |
| Use FusionGAN instead of CycleGAN or DiscoGAN | YouTube Pose image fusion · YouTube Pose user study | identity rating: FusionGAN 3.02; CycleGAN 2.59; DiscoGAN 0.78 | Helped |
| Use FusionGAN instead of CycleGAN or DiscoGAN | YouTube Pose image fusion · YouTube Pose user study | pose rating: FusionGAN 3.06; CycleGAN 2.79; DiscoGAN 0.89 | Helped |
| Use FusionGAN instead of CycleGAN or SimGAN | eye image fusion · Eye dataset user study | identity (reality) rating: FusionGAN 3.97; CycleGAN 2.55; SimGAN 3.02 | Helped |
| Use FusionGAN instead of CycleGAN or SimGAN | eye image fusion · Eye dataset user study | shape (eye gaze) rating: FusionGAN 4.57; CycleGAN 2.63; SimGAN 3.57 | Helped |
| Add the identity loss (ablation) | FusionGAN · qualitative ablation examples | preservation of the second input's shape | Helped |
| Add shape loss LS1 (ablation) | FusionGAN · qualitative ablation examples | training stability and realistic image generation | Helped |
| Add shape losses LS2a and LS2b alongside LS1 (ablation) | FusionGAN · qualitative ablation examples | preservation of the first input's style | Helped |
| Add Min-Patch training (ablation) | FusionGAN · qualitative ablation examples | realism | Helped |
Closest papers by meaning
- Training Generative Adversarial Networks in One Stage (CVPR 2021 · similarity 0.87)
- Training Generative Adversarial Networks from Incomplete Observations using Factorised Discriminators (ICLR 2020 · similarity 0.86)
- Semantically Decomposing the Latent Spaces of Generative Adversarial Networks (ICLR 2018 · similarity 0.86)
- Transformation GAN for Unsupervised Image Synthesis and Representation Learning (CVPR 2020 · similarity 0.86)
- Exact Fusion via Feature Distribution Matching for Few-shot Image Generation (CVPR 2024 · similarity 0.86)
- MCGAN: Enhancing GAN Training with Regression-Based Generator Loss (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.