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

ViTAE: Vision Transformer Advanced by Exploring Intrinsic Inductive Bias

About

Transformers have shown great potential in various computer vision tasks owing to their strong capability in modeling long-range dependency using the self-attention mechanism. Nevertheless, vision transformers treat an image as 1D sequence of visual tokens, lacking an intrinsic inductive bias (IB) in modeling local visual structures and dealing with scale variance. Alternatively, they require large-scale training data and longer training schedules to learn the IB implicitly. In this paper, we propose a novel Vision Transformer Advanced by Exploring intrinsic IB from convolutions, ie, ViTAE. Technically, ViTAE has several spatial pyramid reduction modules to downsample and embed the input image into tokens with rich multi-scale context by using multiple convolutions with different dilation rates. In this way, it acquires an intrinsic scale invariance IB and is able to learn robust feature representation for objects at various scales. Moreover, in each transformer layer, ViTAE has a convolution block in parallel to the multi-head self-attention module, whose features are fused and fed into the feed-forward network. Consequently, it has the intrinsic locality IB and is able to learn local features and global dependencies collaboratively. Experiments on ImageNet as well as downstream tasks prove the superiority of ViTAE over the baseline transformer and concurrent works. Source code and pretrained models will be available at GitHub.

Yufei Xu, Qiming Zhang, Jing Zhang, Dacheng Tao• 2021

Related benchmarks

TaskDatasetResultRank
Semantic segmentationADE20K (val)
mIoU45.4
2731
ClassificationImageNet-1K 1.0 (val)
Top-1 Accuracy (%)82
1155
Instance SegmentationCOCO 2017 (val)--
1144
Semantic segmentationADE20K
mIoU45.4
936
Image ClassificationImageNet-1k (val)
Top-1 Acc82
706
Image ClassificationCIFAR10 (test)
Accuracy98.8
585
Image ClassificationImageNet (val)
Top-1 Accuracy83.6
354
Image ClassificationStanford Cars (test)
Accuracy91.4
306
Image ClassificationCIFAR-100--
302
Instance SegmentationCOCO
APmask42
279
Showing 10 of 31 rows

Other info

Follow for update