Methods › Computer Vision › Image Models › CrossViT
CrossViT
Introduced by Chun-Fu Chen et al. in CrossViT: Cross-Attention Multi-Scale Vision Transformer for Image Classification
archive 2025-07-28 Description, source and code snippet are the archive's method entry.
CrossViT is a type of vision transformer that uses a dual-branch architecture to extract multi-scale feature representations for image classification. The architecture combines image patches (i.e. tokens in a transformer) of different sizes to produce stronger visual features for image classification. It processes small and large patch tokens with two separate branches of different computational complexities and these tokens are fused together multiple times to complement each other.
Fusion is achieved by an efficient cross-attention module, in which each transformer branch creates a non-patch token as an agent to exchange information with the other branch by attention. This allows for linear-time generation of the attention map in fusion instead of quadratic time otherwise.
Papers archive 2025-07-28
3 shown of 3, newest first. Repository counts are the archive's code-links table. A Syntology line states what Syntology ran from that paper's harvested code; it is per sample and not a correctness claim.
-
MaskedFusion360: Reconstruct LiDAR Data by Querying Camera Features 12 Jun 2023 · 1 repository · arXiv:2306.07087
-
Joint rotational invariance and adversarial training of a dual-stream Transformer yields state of the art Brain-Score for Area V4 8 Mar 2022 · 1 repository · arXiv:2203.06649
-
CrossViT: Cross-Attention Multi-Scale Vision Transformer for Image Classification 27 Mar 2021 · 15 repositories · arXiv:2103.14899Syntology ran 17 of 26 samples · 9 unverified
Tasks archive 2025-07-28
5 tasks the archive attaches to papers tagged with this method, by distinct papers. A task without a page in the catalog is plain text.
| Task | Papers |
|---|---|
| Adversarial Robustness | 1 |
| General Classification | 1 |
| Image Classification | 1 |
| Sensor Fusion | 1 |
| image-classification | 1 |
Usage over time archive 2025-07-28
Components: the archive holds no method-to-method composition, so PwC's Components table cannot be rebuilt; the Papers list carries no Results column for the same reason (the archive does not join its leaderboard rows to method tags).
Categories archive 2025-07-28
Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections