# Instance Segmentation with Point Supervision

arXiv: https://arxiv.org/abs/1906.06392 · HTML: https://mlsanity.com/paper/1906.06392/

## Key facts

- Cited by 8 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 14 Jun 2019.

## Built on by (most cited first, 8 of 8)

- [BBAM: Bounding Box Attribution Map for Weakly Supervised Semantic and Instance Segmentation](https://mlsanity.com/paper/2103.08907/) (2021)
- [A Survey of Self-Supervised and Few-Shot Object Detection](https://mlsanity.com/paper/2110.14711/) (2021)
- [From colouring-in to pointillism: revisiting semantic segmentation supervision](https://mlsanity.com/paper/2210.14142/) (2022)
- [Weakly-supervised Instance Segmentation via Class-agnostic Learning with Salient Images](https://mlsanity.com/paper/2104.01526/) (2021)
- [A Deep Learning Localization Method for Measuring Abdominal Muscle Dimensions in Ultrasound Images](https://mlsanity.com/paper/2109.14919/) (2021)
- [Spatial Frequency Modulation for Semantic Segmentation](https://mlsanity.com/paper/2507.11893/) (2025)
- [Points2Polygons: Context-Based Segmentation from Weak Labels Using Adversarial Networks](https://mlsanity.com/paper/2106.02804/) (2021)
- [Dilation-Erosion for Single-Frame Supervised Temporal Action Localization](https://mlsanity.com/paper/2212.06348/) (2022)

## Abstract (arXiv)

Instance segmentation methods often require costly per-pixel labels. We propose a method that only requires point-level annotations. During training, the model only has access to a single pixel label per object, yet the task is to output full segmentation masks. To address this challenge, we construct a network with two branches: (1) a localization network (L-Net) that predicts the location of each object; and (2) an embedding network (E-Net) that learns an embedding space where pixels of the same object are close. The segmentation masks for the located objects are obtained by grouping pixels with similar embeddings. At training time, while L-Net only requires point-level annotations, E-Net uses pseudo-labels generated by a class-agnostic object proposal method. We evaluate our approach on PASCAL VOC, COCO, KITTI and CityScapes datasets. The experiments show that our method (1) obtains competitive results compared to fully-supervised methods in certain scenarios; (2) outperforms fully- and weakly- supervised methods with a fixed annotation budget; and (3) is a first strong baseline for instance segmentation with point-level supervision.