# Superhuman Accuracy on the SNEMI3D Connectomics Challenge

arXiv: https://arxiv.org/abs/1706.00120 · HTML: https://mlsanity.com/paper/1706.00120/

## Key facts

- Cited by 64 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 31 May 2017.

## Built on by (most cited first, 30 of 64)

- [Harmonizing the object recognition strategies of deep neural networks with humans](https://mlsanity.com/paper/2211.04533/) (2022)
- [Deep Implicit Statistical Shape Models for 3D Medical Image Delineation](https://mlsanity.com/paper/2104.02847/) (2021)
- [Self-Supervised Neuron Segmentation with Multi-Agent Reinforcement Learning](https://mlsanity.com/paper/2310.04148/) (2023)
- [Learning Multiscale Consistency for Self-supervised Electron Microscopy Instance Segmentation](https://mlsanity.com/paper/2308.09917/) (2023)
- [The 3D-PC: a benchmark for visual perspective taking in humans and machines](https://mlsanity.com/paper/2406.04138/) (2024)
- [Understanding Visual Feature Reliance through the Lens of Complexity](https://mlsanity.com/paper/2407.06076/) (2024)
- [Deep Rib Fracture Instance Segmentation and Classification from CT on the RibFrac Challenge](https://mlsanity.com/paper/2402.09372/) (2024)
- [Deep Lake: a Lakehouse for Deep Learning](https://mlsanity.com/paper/2209.10785/) (2022)
- [NucMM Dataset: 3D Neuronal Nuclei Instance Segmentation at Sub-Cubic Millimeter Scale](https://mlsanity.com/paper/2107.05840/) (2021)
- [AxonEM Dataset: 3D Axon Instance Segmentation of Brain Cortical Regions](https://mlsanity.com/paper/2107.05451/) (2021)
- [Sparse Object-level Supervision for Instance Segmentation with Pixel Embeddings](https://mlsanity.com/paper/2103.14572/) (2021)
- [Advanced Deep Networks for 3D Mitochondria Instance Segmentation](https://mlsanity.com/paper/2104.07961/) (2021)
- [An Analysis on Ensemble Learning optimized Medical Image Classification with Deep Convolutional Neural Networks](https://mlsanity.com/paper/2201.11440/) (2022)
- [GOV-NeSF: Generalizable Open-Vocabulary Neural Semantic Fields](https://mlsanity.com/paper/2404.00931/) (2024)
- [How Shift Equivariance Impacts Metric Learning for Instance Segmentation](https://mlsanity.com/paper/2101.05846/) (2021)
- [MTNeuro: A Benchmark for Evaluating Representations of Brain Structure Across Multiple Levels of Abstraction](https://mlsanity.com/paper/2301.00345/) (2023)
- [TokenUnify: Scaling Up Autoregressive Pretraining for Neuron Segmentation](https://mlsanity.com/paper/2405.16847/) (2024)
- [Does DINOv3 Set a New Medical Vision Standard? Benchmarking 2D and 3D Classification, Segmentation, and Registration](https://mlsanity.com/paper/2509.06467/) (2025)
- [Segmentation in large-scale cellular electron microscopy with deep learning: A literature survey](https://mlsanity.com/paper/2206.07171/) (2022)
- [Self-supervised learning via inter-modal reconstruction and feature projection networks for label-efficient 3D-to-2D segmentation](https://mlsanity.com/paper/2307.03008/) (2023)
- [Automatic Volumetric Segmentation of Additive Manufacturing Defects with 3D U-Net](https://mlsanity.com/paper/2101.08993/) (2021)
- [Asymmetry Disentanglement Network for Interpretable Acute Ischemic Stroke Infarct Segmentation in Non-Contrast CT Scans](https://mlsanity.com/paper/2206.15445/) (2022)
- [Learning Multimodal Volumetric Features for Large-Scale Neuron Tracing](https://mlsanity.com/paper/2401.03043/) (2024)
- [HResFormer: Hybrid Residual Transformer for Volumetric Medical Image Segmentation](https://mlsanity.com/paper/2412.11458/) (2024)
- [H3DE-Net: Efficient and Accurate 3D Landmark Detection in Medical Imaging](https://mlsanity.com/paper/2502.14221/) (2025)
- [Consistent Recurrent Neural Networks for 3D Neuron Segmentation](https://mlsanity.com/paper/2102.01021/) (2021)
- [Voxel-level Siamese Representation Learning for Abdominal Multi-Organ Segmentation](https://mlsanity.com/paper/2105.07672/) (2021)
- [PyTorch Connectomics: A Scalable and Flexible Segmentation Framework for EM Connectomics](https://mlsanity.com/paper/2112.05754/) (2021)
- [Instance Segmentation of Unlabeled Modalities via Cyclic Segmentation GAN](https://mlsanity.com/paper/2204.03082/) (2022)
- [Scan2Part: Fine-grained and Hierarchical Part-level Understanding of Real-World 3D Scans](https://mlsanity.com/paper/2206.02366/) (2022)

## Abstract (arXiv)

For the past decade, convolutional networks have been used for 3D reconstruction of neurons from electron microscopic (EM) brain images. Recent years have seen great improvements in accuracy, as evidenced by submissions to the SNEMI3D benchmark challenge. Here we report the first submission to surpass the estimate of human accuracy provided by the SNEMI3D leaderboard. A variant of 3D U-Net is trained on a primary task of predicting affinities between nearest neighbor voxels, and an auxiliary task of predicting long-range affinities. The training data is augmented by simulated image defects. The nearest neighbor affinities are used to create an oversegmentation, and then supervoxels are greedily agglomerated based on mean affinity. The resulting SNEMI3D score exceeds the estimate of human accuracy by a large margin. While one should be cautious about extrapolating from the SNEMI3D benchmark to real-world accuracy of large-scale neural circuit reconstruction, our result inspires optimism that the goal of full automation may be realizable in the future.