Vision Transformers Need Registers

• URL: http://arxiv.org/abs/2309.16588v2
• Date: Fri, 12 Apr 2024 09:38:33 GMT
• Title: Vision Transformers Need Registers
• Authors: Timothée Darcet, Maxime Oquab, Julien Mairal, Piotr Bojanowski,
• Abstract summary: We identify and characterize artifacts in feature maps of both supervised and self-supervised ViT networks. We show that this solution fixes that problem entirely for both supervised and self-supervised models.
• Score: 26.63912173005165
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Transformers have recently emerged as a powerful tool for learning visual representations. In this paper, we identify and characterize artifacts in feature maps of both supervised and self-supervised ViT networks. The artifacts correspond to high-norm tokens appearing during inference primarily in low-informative background areas of images, that are repurposed for internal computations. We propose a simple yet effective solution based on providing additional tokens to the input sequence of the Vision Transformer to fill that role. We show that this solution fixes that problem entirely for both supervised and self-supervised models, sets a new state of the art for self-supervised visual models on dense visual prediction tasks, enables object discovery methods with larger models, and most importantly leads to smoother feature maps and attention maps for downstream visual processing.

Related papers

• Learning Visual Prompts for Guiding the Attention of Vision Transformers [29.023448448047805]
  This work proposes to learn the visual prompts for guiding the attention of vision transformers. The learned visual prompt, added to any input image would redirect the attention of the pre-trained vision transformer to its spatial location on the image.
  arXiv  Detail & Related papers  (2024-06-05T14:13:38Z)
• Analyzing Local Representations of Self-supervised Vision Transformers [34.56680159632432]
  We present a comparative analysis of various self-supervised Vision Transformers (ViTs) Inspired by large language models, we examine the abilities of ViTs to perform various computer vision tasks with little to no fine-tuning.
  arXiv  Detail & Related papers  (2023-12-31T11:38:50Z)
• Attention Deficit is Ordered! Fooling Deformable Vision Transformers with Collaborative Adversarial Patches [3.4673556247932225]
  Deformable vision transformers significantly reduce the complexity of attention modeling. Recent work has demonstrated adversarial attacks against conventional vision transformers. We develop new collaborative attacks where a source patch manipulates attention to point to a target patch, which contains the adversarial noise to fool the model.
  arXiv  Detail & Related papers  (2023-11-21T17:55:46Z)
• VISIT: Visualizing and Interpreting the Semantic Information Flow of Transformers [45.42482446288144]
  Recent advances in interpretability suggest we can project weights and hidden states of transformer-based language models to their vocabulary. We investigate LM attention heads and memory values, the vectors the models dynamically create and recall while processing a given input. We create a tool to visualize a forward pass of Generative Pre-trained Transformers (GPTs) as an interactive flow graph.
  arXiv  Detail & Related papers  (2023-05-22T19:04:56Z)
• AttentionViz: A Global View of Transformer Attention [60.82904477362676]
  We present a new visualization technique designed to help researchers understand the self-attention mechanism in transformers. The main idea behind our method is to visualize a joint embedding of the query and key vectors used by transformer models to compute attention. We create an interactive visualization tool, AttentionViz, based on these joint query-key embeddings.
  arXiv  Detail & Related papers  (2023-05-04T23:46:49Z)
• Learning Explicit Object-Centric Representations with Vision Transformers [81.38804205212425]
  We build on the self-supervision task of masked autoencoding and explore its effectiveness for learning object-centric representations with transformers. We show that the model efficiently learns to decompose simple scenes as measured by segmentation metrics on several multi-object benchmarks.
  arXiv  Detail & Related papers  (2022-10-25T16:39:49Z)
• Optimizing Relevance Maps of Vision Transformers Improves Robustness [91.61353418331244]
  It has been observed that visual classification models often rely mostly on the image background, neglecting the foreground, which hurts their robustness to distribution changes. We propose to monitor the model's relevancy signal and manipulate it such that the model is focused on the foreground object. This is done as a finetuning step, involving relatively few samples consisting of pairs of images and their associated foreground masks.
  arXiv  Detail & Related papers  (2022-06-02T17:24:48Z)
• PreViTS: Contrastive Pretraining with Video Tracking Supervision [53.73237606312024]
  PreViTS is an unsupervised SSL framework for selecting clips containing the same object. PreViTS spatially constrains the frame regions to learn from and trains the model to locate meaningful objects. We train a momentum contrastive (MoCo) encoder on VGG-Sound and Kinetics-400 datasets with PreViTS.
  arXiv  Detail & Related papers  (2021-12-01T19:49:57Z)
• Efficient Self-supervised Vision Transformers for Representation Learning [86.57557009109411]
  We show that multi-stage architectures with sparse self-attentions can significantly reduce modeling complexity. We propose a new pre-training task of region matching which allows the model to capture fine-grained region dependencies. Our results show that combining the two techniques, EsViT achieves 81.3% top-1 on the ImageNet linear probe evaluation.
  arXiv  Detail & Related papers  (2021-06-17T19:57:33Z)
• VinVL: Revisiting Visual Representations in Vision-Language Models [96.39332942534368]
  We develop an improved object detection model to provide object-centric representations of images. New visual features significantly improve the performance across all vision language (VL) tasks. We will release the new object detection model to public.
  arXiv  Detail & Related papers  (2021-01-02T23:35:27Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.