# Self-Supervised Pre-training of Vision Transformers for Dense Prediction Tasks

Source: [Self-Supervised Pre-training of Vision Transformers for Dense Prediction Tasks](<https://www.sota2.com/research/paper/68fa6cb41e209131cea355cc>)

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

Data updated: 2025-10-26T00:16:46.056Z

Authors: Jaonary Rabarisoa, Valentin Belissen, Florian Chabot, Quoc-Cuong Pham

Published: 2022-05-30T15:25:37.000Z

## Abstract

We present a new self-supervised pre-training of Vision Transformers for dense prediction tasks. It is based on a contrastive loss across views that compares pixel-level representations to global image representations. This strategy produces better local features suitable for dense prediction tasks as opposed to contrastive pre-training based on global image representation only. Furthermore, our approach does not suffer from a reduced batch size since the number of negative examples needed in the contrastive loss is in the order of the number of local features. We demonstrate the effectiveness of our pre-training strategy on two dense prediction tasks: semantic segmentation and monocular depth estimation.

## 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.

## Original sources and implementation

[Original paper](<https://arxiv.org/abs/2205.15173>)

[PDF](<https://arxiv.org/pdf/2205.15173.pdf>)