Your ViT is Secretly an Image Segmentation Model

IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings - IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings, pp. 25303–25313

Abstract

Vision Transformers (ViTs) have shown remarkable performance and scalability across various computer vision tasks. To apply single-scale ViTs to image segmentation, existing methods adopt a convolutional adapter to generate multi-scale features, a pixel decoder to fuse these features, and a Transformer decoder that uses the fused features to make predictions. In this paper, we show that the inductive biases introduced by these task-specific components can instead be learned by the ViT itself, given sufficiently large models and extensive pre-training. Based on these findings, we introduce the Encoder-Only Mask Transformer (EoMT), which repurposes the plain ViT architecture to conduct image segmentation. With large-scale models and pre-training, EoMT obtains a segmentation accuracy similar to state-of-the-art models that use task-specific components. At the same time, EoMT is significantly faster than these methods due to its architectural simplicity, e.g., up to 4 × faster with ViT-L. Across a range of model sizes, EoMT demonstrates an optimal balance between segmentation accuracy and prediction speed, suggesting that compute resources are better spent on scaling the ViT itself rather than adding architectural complexity. Code: https://www.tue-mps.org/eomt/.

Authors 8

1. Tommie Kerssies

   Eindhoven University of Technology

   Affiliation as printed

   Eindhoven University of Technology

2. Niccolò Cavagnero

   Politecnico di Torino

   Affiliation as printed

   Polytechnic of Turin

3. Alexander Hermans Aachen

   RWTH Aachen University

   Affiliation as printed

   RWTH Aachen University

4. Narges Norouzi

   Eindhoven University of Technology

   Affiliation as printed

   Eindhoven University of Technology

5. Giuseppe Averta

   Politecnico di Torino

   Affiliation as printed

   Polytechnic of Turin

6. Bastian Leibe Aachen

   RWTH Aachen University

   Affiliation as printed

   RWTH Aachen University

7. Gijs Dubbelman

   Eindhoven University of Technology

   Affiliation as printed

   Eindhoven University of Technology

8. Daan de Geus

   Eindhoven University of Technology

   Affiliation as printed

   Eindhoven University of Technology