Research map / 3D vision / Depth and stereo
NIS3D: A Completely Annotated Benchmark for Dense 3D Nuclei Image Segmentation
Wei Zheng, James Cheng Peng, Zeyuan Hou, Boyu Lyu, Mengfan Wang, Xuelong Mi, Shuoxuan Qiao, Yinan Wan, Guoqiang Yu
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 |
|---|---|---|---|
| Compare StarDist with Human annotators | StarDist · NIS3D Zebrafish 1 | W-SEG: 0.192 vs 0.905 | Hurt |
| Use StarDist instead of 3D Suite iterative thresholding | StarDist · NIS3D Zebrafish 1 | W-F1: 0.477 -> 0.586 | Improved |
| Use StarDist instead of 3D Suite iterative thresholding | StarDist · NIS3D Drosophila 2 | W-F1: 0.877 -> 0.684 | Hurt |
| Use StarDist instead of 3D Suite iterative thresholding | StarDist · NIS3D Mus Musculus 2 | W-F1: 0.667 -> 0.540 | Hurt |
| Use Cellpose cyto model instead of nuclei model | Cellpose · NIS3D Zebrafish 1 | W-F1: 0.492 -> 0.098 | Hurt |
| Use Cellpose cyto model instead of nuclei model | Cellpose · NIS3D Zebrafish 2 | W-F1: 0.654 -> 0.661 | Improved |
| Use Cellpose cyto model instead of nuclei model | Cellpose · NIS3D Mus Musculus 1 | W-F1: 0.356 -> 0.536 | Improved |
| Use Cellpose cyto model instead of nuclei model | Cellpose · NIS3D Drosophila 2 | W-F1: 0.278 -> 0.092 | Hurt |
| Evaluate Vaa3D on foreground versus instance quality | Vaa3D · NIS3D Zebrafish 2 | W-IoU vs W-SEG: 0.703 vs 0.007 | Mixed |
| Use QCAnet on Zebrafish 2 versus Zebrafish 1 | QCAnet · NIS3D | W-F1: 0.510 -> 0.031 | Hurt |
| Human annotation on small-cell Zebrafish 2 versus other data | Human · NIS3D Zebrafish 2 | W-SEG: 0.905 -> 0.610 | Hurt |
| Cellpose nuclei model on Drosophila 1 precision | Cellpose_nuclei · NIS3D Drosophila 1 | W-Prec: 0.457 vs 0.881 | Hurt |
Closest papers by meaning
- CiNuSeg: Class Incremental Nuclei Segmentation via Anchor-driven Consistency Learning with Dual Region Regularization (AAAI 2026 · similarity 0.86)
- Superadditivity and Convex Optimization for Globally Optimal Cell Segmentation Using Deformable Shape Models (TPAMI 2022 · similarity 0.86)
- Unleashing the Power of Prompt-driven Nucleus Instance Segmentation (ECCV 2024 · similarity 0.85)
- Harnessing SSL for Segmentation in 3D Microscopy with Noisy Labels and Hard Patches (ECCV 2026 · similarity 0.85)
- VISTA3D: A Unified Segmentation Foundation Model For 3D Medical Imaging (CVPR 2025 · similarity 0.84)
- Mudslide: A Universal Nuclear Instance Segmentation Method (CVPR 2024 · similarity 0.84)
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.
