3DV's most-cited paper is medical V-Net, source of the Dice loss
ducha_aiki · x · 2026-10-20
Per an official 3DV countdown, the conference's most-cited paper is the medical imaging work V-Net, with more citations than the other four papers on the list combined. V-Net segmented MRI volumes with a 3D fully convolutional network and introduced the Dice loss to handle class imbalance — the loss function widely used in segmentation today originated there.
More from Research
- Mirror lets agents follow video-model plans, crafting an iron pickaxe in Minecraft — vincesitzmann · 2026-10-20
- Scenario releases 1,342 high-quality AI-generated 3D assets as open CC BY 4.0 dataset — victormustar · 2026-10-20
- Scholar writes AI-assisted book on storytelling values, work nominated for EMNLP best paper — _akpiper · 2026-10-20
- Google's T2J-Bench shows coding agents score just 0.676 migrating PyTorch repos to JAX — CShorten30 · 2026-10-20
- New proof: perfect reconstruction can't guarantee VAE latent codes read back as written — Hidenori8Tanaka · 2026-10-20
- Math model shows when errors tip into epidemic spread across LLM agent swarms — CourseSome8119 · 2026-10-20