Skip to Main content Skip to Navigation
Conference papers

Autoregressive Unsupervised Image Segmentation

Abstract : In this work, we propose a new unsupervised image segmentation approach based on mutual information maximization between different constructed views of the inputs. Taking inspiration from autoregressive generative models that predict the current pixel from past pixels in a raster-scan ordering created with masked convolutions, we propose to use different orderings over the inputs using various forms of masked convolutions to construct different views of the data. For a given input, the model produces a pair of predictions with two valid orderings, and is then trained to maximize the mutual information between the two outputs. These outputs can either be low-dimensional features for representation learning or output clusters corresponding to semantic labels for clustering. While masked convolutions are used during training, in inference, no masking is applied and we fall back to the standard convolution where the model has access to the full input. The proposed method outperforms current state-of-the-art on unsupervised image segmentation. It is simple and easy to implement, and can be extended to other visual tasks and integrated seamlessly into existing unsupervised learning methods requiring different views of the data.
Document type :
Conference papers
Complete list of metadata
Contributor : Myriam Tami Connect in order to contact the contributor
Submitted on : Thursday, December 10, 2020 - 10:01:25 AM
Last modification on : Tuesday, December 14, 2021 - 3:01:54 AM
Long-term archiving on: : Thursday, March 11, 2021 - 6:52:53 PM


Files produced by the author(s)



yassine Ouali, Céline Hudelot, Myriam Tami. Autoregressive Unsupervised Image Segmentation. ECCV 2020, Aug 2020, Glasgow (on line), United Kingdom. ⟨10.1007/978-3-030-58571-6_9⟩. ⟨hal-03050599⟩



Record views


Files downloads