# Affinity Derivation and Graph Merge for Instance Segmentation

Source: [Affinity Derivation and Graph Merge for Instance Segmentation](<https://www.sota2.com/research/paper/68fa651d1e209131cea08772>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2025-10-28T13:56:37.466Z

Authors: Yiding Liu, Siyu Yang, Bin Li, Wengang Zhou, Jizheng Xu, Houqiang Li, Yan Lu

Published: 2018-11-27T08:34:28.000Z

## Abstract

We present an instance segmentation scheme based on pixel affinity information, which is the relationship of two pixels belonging to a same instance. In our scheme, we use two neural networks with similar structure. One is to predict pixel level semantic score and the other is designed to derive pixel affinities.   Regarding pixels as the vertexes and affinities as edges, we then propose a simple yet effective graph merge algorithm to cluster pixels into instances. Experimental results show that our scheme can generate fine-grained instance mask.   With Cityscapes training data, the proposed scheme achieves 27.3 AP on test set.

## Recorded benchmark results

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

- [Instance Segmentation on iShape](<https://www.sota2.com/research/sota/instance-segmentation-on-ishape.md>): Instance Segmentation; iShape; mmAP (Average): 41.21.

- [Instance Segmentation on Cityscapes (test)](<https://www.sota2.com/research/sota/instance-segmentation-on-cityscapes-test.md>): Instance Segmentation; Cityscapes (test); AP (Overall): 27.6.

## Original sources and implementation

[Original paper](<https://arxiv.org/abs/1811.10870>)

[PDF](<https://arxiv.org/pdf/1811.10870.pdf>)
