Our new X account is live! Follow @wizwand_team for updates
WorkDL logo mark

Vision Transformers for Dense Prediction

About

We introduce dense vision transformers, an architecture that leverages vision transformers in place of convolutional networks as a backbone for dense prediction tasks. We assemble tokens from various stages of the vision transformer into image-like representations at various resolutions and progressively combine them into full-resolution predictions using a convolutional decoder. The transformer backbone processes representations at a constant and relatively high resolution and has a global receptive field at every stage. These properties allow the dense vision transformer to provide finer-grained and more globally coherent predictions when compared to fully-convolutional networks. Our experiments show that this architecture yields substantial improvements on dense prediction tasks, especially when a large amount of training data is available. For monocular depth estimation, we observe an improvement of up to 28% in relative performance when compared to a state-of-the-art fully-convolutional network. When applied to semantic segmentation, dense vision transformers set a new state of the art on ADE20K with 49.02% mIoU. We further show that the architecture can be fine-tuned on smaller datasets such as NYUv2, KITTI, and Pascal Context where it also sets the new state of the art. Our models are available at https://github.com/intel-isl/DPT.

Ren\'e Ranftl, Alexey Bochkovskiy, Vladlen Koltun• 2021

Related benchmarks

TaskDatasetResultRank
Semantic segmentationADE20K (val)
mIoU49.02
2731
Semantic segmentationADE20K
mIoU49.2
936
Monocular Depth EstimationKITTI (Eigen)
Abs Rel0.062
502
Depth EstimationNYU v2 (test)
Threshold Accuracy (delta < 1.25)90.4
423
Semantic segmentationPASCAL Context (val)
mIoU60.5
323
Depth EstimationKITTI (Eigen split)
RMSE2.573
276
Monocular Depth EstimationNYU v2 (test)
Abs Rel0.094
257
Monocular Depth EstimationKITTI (Eigen split)
Abs Rel0.052
193
Depth EstimationNYU Depth V2--
177
Semantic segmentationPascal Context (test)
mIoU60.46
176
Showing 10 of 107 rows
...

Other info

Code

Follow for update