# Learning the Superpixel in a Non-iterative and Lifelong Manner

arXiv: https://arxiv.org/abs/2103.10681 · HTML: https://mlsanity.com/paper/2103.10681/

## Key facts

- Cited by 12 papers in the map (data through 14 Oct 2026).
- Builds on 8 of its 31 references that are in the map.
- First posted on arXiv 19 Mar 2021.

## Builds on (most cited first, 8 of 8)

- [Adam: A Method for Stochastic Optimization](https://mlsanity.com/paper/1412.6980/) (2014)
- [DeepLab: Semantic Image Segmentation with Deep Convolutional Nets, Atrous Convolution, and Fully Connected CRFs](https://mlsanity.com/paper/1606.00915/) (2016)
- [Unsupervised Deep Embedding for Clustering Analysis](https://mlsanity.com/paper/1511.06335/) (2015)
- [Superpixel Sampling Networks](https://mlsanity.com/paper/1807.10174/) (2018)
- [SEEDS: Superpixels Extracted via Energy-Driven Sampling](https://mlsanity.com/paper/1309.3848/) (2013)
- [Superpixel Segmentation with Fully Convolutional Networks](https://mlsanity.com/paper/2003.12929/) (2020)
- [OpenLORIS-Object: A Robotic Vision Dataset and Benchmark for Lifelong Deep Learning](https://mlsanity.com/paper/1911.06487/) (2019)
- [Superpixel Segmentation via Convolutional Neural Networks with Regularized Information Maximization](https://mlsanity.com/paper/2002.06765/) (2020)

## Built on by (most cited first, 12 of 12)

- [Superpoint Gaussian Splatting for Real-Time High-Fidelity Dynamic Scene Reconstruction](https://mlsanity.com/paper/2406.03697/) (2024)
- [CLUSTSEG: Clustering for Universal Segmentation](https://mlsanity.com/paper/2305.02187/) (2023)
- [Transformers in 3D Point Clouds: A Survey](https://mlsanity.com/paper/2205.07417/) (2022)
- [A comprehensive review and new taxonomy on superpixel segmentation](https://mlsanity.com/paper/2409.19179/) (2024)
- [Interactive Cartoonization with Controllable Perceptual Factors](https://mlsanity.com/paper/2212.09555/) (2022)
- [SuperCL: Superpixel Guided Contrastive Learning for Medical Image Segmentation Pre-training](https://mlsanity.com/paper/2504.14737/) (2025)
- [Superpixel Anything: A general object-based framework for accurate yet regular superpixel segmentation](https://mlsanity.com/paper/2509.12791/) (2025)
- [Superpixel Segmentation: A Long-Lasting Ill-Posed Problem](https://mlsanity.com/paper/2411.06478/) (2024)
- [Improve Retinal Artery/Vein Classification via Channel Couplin](https://mlsanity.com/paper/2508.03738/) (2025)
- [Unsupervised Superpixel Generation using Edge-Sparse Embedding](https://mlsanity.com/paper/2211.15474/) (2022)
- [SuperpixelGraph: Semi-automatic generation of building footprint through semantic-sensitive superpixel and neural graph networks](https://mlsanity.com/paper/2304.05661/) (2023)
- [Computer Vision with a Superpixelation Camera](https://mlsanity.com/paper/2603.26900/) (2026)

## Abstract (arXiv)

Superpixel is generated by automatically clustering pixels in an image into hundreds of compact partitions, which is widely used to perceive the object contours for its excellent contour adherence. Although some works use the Convolution Neural Network (CNN) to generate high-quality superpixel, we challenge the design principles of these networks, specifically for their dependence on manual labels and excess computation resources, which limits their flexibility compared with the traditional unsupervised segmentation methods. We target at redefining the CNN-based superpixel segmentation as a lifelong clustering task and propose an unsupervised CNN-based method called LNS-Net. The LNS-Net can learn superpixel in a non-iterative and lifelong manner without any manual labels. Specifically, a lightweight feature embedder is proposed for LNS-Net to efficiently generate the cluster-friendly features. With those features, seed nodes can be automatically assigned to cluster pixels in a non-iterative way. Additionally, our LNS-Net can adapt the sequentially lifelong learning by rescaling the gradient of weight based on both channel and spatial context to avoid overfitting. Experiments show that the proposed LNS-Net achieves significantly better performance on three benchmarks with nearly ten times lower complexity compared with other state-of-the-art methods.