Research map / Image generation / Editing and style transfer
High-Resolution Daytime Translation Without Domain Labels
Ivan Anokhin, Pavel Solovev, Denis Korzhenkov, Alexey Kharlamov, Taras Khakhulin, Aleksei Silvestrov, Sergey Nikolenko, Victor Lempitsky, Gleb Sterkin
Check 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 |
|---|---|---|---|
| Train without daytime domain labels | HiDT · Daytime landscape translation | Human preference and GAN metrics | No clear effect |
| Use HiDT instead of labeled baselines | Daytime translation · Hold-out dataset | DIPD, CIS, and Inception score: DIPD: 0.691; CIS: 1.559; IS random: 1.673; IS swapped: 1.605 | Mixed |
| Increase the number of averaged styles from N=1 to N=10 | FUNIT-T · Daytime translation user study | HiDT user preference score: 0.51; 0.48; 0.47 | Hurt |
| Add the projection conditioning scheme to the discriminator | HiDT · Daytime landscape translation | Translation quality | Improved |
| Add auxiliary segmentation loss (ablation) | HiDT · Daytime landscape translation | Semantic consistency | Improved |
| Remove auxiliary segmentation loss (ablation) | HiDT · Daytime landscape translation | User preference score | No clear effect |
| Use the autoencoding loss alone (ablation) | HiDT · Hold-out dataset, swapped styles | CIS: 1.559 -> 1.179 | Hurt |
| Use the autoencoding loss alone (ablation) | HiDT · Daytime style transfer | Target daytime resemblance | Hurt |
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.
- DRIT — DRIT++: Diverse Image-to-Image Translation via Disentangled Representations (IJCV 2020) · better on 5 of 5 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
- DNIT: Enhancing Day-Night Image-to-Image Translation through Fine-Grained Feature Handling (Student Abstract) (AAAI 2024 · similarity 0.86)
- TraVeLGAN: Image-To-Image Translation by Transformation Vector Learning (CVPR 2019 · similarity 0.85)
- Discriminative Region Proposal Adversarial Networks for High-Quality Image-to-Image Translation (ECCV 2018 · similarity 0.85)
- Discriminative Region Proposal Adversarial Network for High-Quality Image-to-Image Translation (IJCV 2019 · similarity 0.85)
- Rethinking Nighttime Image Deraining via Learnable Color Space Transformation (NeurIPS 2025 · similarity 0.83)
- Text2Relight: Creative Portrait Relighting with Text Guidance (AAAI 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.
