Research map / Image generation / GANs
Escaping from Collapsing Modes in a Constrained Space
Chia-Che Chang, Chieh Hubert Lin, Che-Rung Lee, Da-Cheng Juan, Wei Wei, Hwann-Tzong Chen
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 latent-space constraint loss to the discriminator of BEGAN | BEGAN · CelebA 64x64 (full) | epoch of mode collapse: collapse at 25th epoch -> no collapse up to 100 epochs | Improved |
| add latent-space constraint loss to BEGAN on a 1/10 sized CelebA subset | BEGAN · CelebA (1/10 subset) | training stability and image quality (qualitative) | Improved |
| add constraint loss so Enc(x) approximates z* in one shot instead of z*-search | BEGAN · CelebA latent inversion | inversion time and fidelity (qualitative): z*-search 10,000 iterations takes 340 seconds to 3,970 seconds vs one encoder pass | Improved |
Closest papers by meaning
- Seeing What a GAN Cannot Generate (ICCV 2019 · similarity 0.89)
- DynGAN: Solving Mode Collapse in GANs With Dynamic Clustering (TPAMI 2024 · similarity 0.89)
- PacGAN: The power of two samples in generative adversarial networks (NeurIPS 2018 · similarity 0.88)
- VEEGAN: Reducing Mode Collapse in GANs using Implicit Variational Learning (NeurIPS 2017 · similarity 0.88)
- Complexity Matters: Rethinking the Latent Space for Generative Modeling (NeurIPS 2023 · similarity 0.88)
- Combating Mode Collapse via Offline Manifold Entropy Estimation (AAAI 2023 · similarity 0.88)
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.
