# What's the Point: Semantic Segmentation with Point Supervision

Source: [What's the Point: Semantic Segmentation with Point Supervision](<https://www.sota2.com/research/paper/68fa62a31e209131ce9fa6fd>)

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

Data updated: 2025-10-27T05:50:53.268Z

Authors: Amy Bearman, Olga Russakovsky, Vittorio Ferrari, Li Fei-Fei

Published: 2015-06-06T02:45:48.000Z

## Abstract

The semantic image segmentation task presents a trade-off between test time accuracy and training-time annotation cost. Detailed per-pixel annotations enable training accurate models but are very time-consuming to obtain, image-level class labels are an order of magnitude cheaper but result in less accurate models. We take a natural step from image-level annotation towards stronger supervision: we ask annotators to point to an object if one exists. We incorporate this point supervision along with a novel objectness potential in the training loss function of a CNN model. Experimental results on the PASCAL VOC 2012 benchmark reveal that the combined effect of point-level supervision and objectness potential yields an improvement of 12.9% mIOU over image-level supervision. Further, we demonstrate that models trained with point-level supervision are more accurate than models trained with image-level, squiggle-level or full supervision given a fixed annotation budget.

## 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.

- [Semantic Segmentation on OCT (test)](<https://www.sota2.com/research/sota/semantic-segmentation-on-oct-test.md>): Semantic segmentation; OCT (test); Relative Performance: 95.4.

- [Semantic Segmentation on VOC (test)](<https://www.sota2.com/research/sota/semantic-segmentation-on-voc-test.md>): Semantic segmentation; VOC (test); Relative Performance: 46.

- [Semantic Segmentation on SUIM (test)](<https://www.sota2.com/research/sota/semantic-segmentation-on-suim-test.md>): Semantic segmentation; SUIM (test); Relative Performance: 81.3.

- [Semantic Segmentation on PASCAL VOC 2012 (val)](<https://www.sota2.com/research/sota/semantic-segmentation-on-pascal-voc-2012-val.md>): Semantic segmentation; PASCAL VOC 2012 (val); Mean IoU: 45.1.

- [Semantic Segmentation on Cityscapes (test)](<https://www.sota2.com/research/sota/semantic-segmentation-on-cityscapes-test.md>): Semantic segmentation; Cityscapes (test); mIoU: not recorded.

## Original sources and implementation

[Original paper](<https://arxiv.org/abs/1506.02106>)

[PDF](<https://arxiv.org/pdf/1506.02106.pdf>)