Text-to-image generation has traditionally focused on finding better modeling assumptions for training on a fixed dataset. These assumptions might involve complex architectures, auxiliary losses, or side information such as object part labels or segmentation masks supplied during training. We describe a simple approach for this task based on a transformer that autoregressively models the text and image tokens as a single stream of data. With sufficient data and scale, our approach is competitive with previous domain-specific models when evaluated in a zero-shot fashion.
The recent explosive interest on transformers has suggested their potential to become powerful “universal” models for computer vision tasks, such as classification, detection, and segmentation. However, how further transformers can go – are they ready to take some more notoriously difficult vision tasks, e.g., generative adversarial networks (GANs)?Driven by that curiosity, we conduct the first pilot study in building a GAN completely free of convolutions, using only pure transformer-based architectures. Our vanilla GAN architecture, dubbed TransGAN, consists of a memory-friendly transformer-based generator that progressively increases feature resolution while decreasing embedding dimension, and a patch-level discriminator that is also transformer-based. We then demonstrate TransGAN to notably benefit from data augmentations (more than standard GANs), a multi-task co-training strategy for the generator, and a locally initialized self-attention that emphasizes the neighborhood smoothness of natural images. Equipped with those findings, TransGAN can effectively scale up with bigger models and high-resolution image datasets. Specifically, our best architecture achieves highly competitive performance compared to current state-of-the-art GANs based on convolutional backbones. Specifically, TransGAN sets new state-of-the-art IS score of 10.10 and FID score of 25.32 on STL-10. It also reaches competitive 8.64 IS score and 11.89 FID score on Cifar-10, and 12.23 FID score on CelebA 64×64, respectively. We also conclude with a discussion of the current limitations and future potential of TransGAN.
In this paper, we explore the Vision Transformer (ViT), a pure transformer-based model, for the object re-identification (ReID) task. With several adaptations, a strong baseline ViT-BoT is constructed with ViT as backbone, which achieves comparable results to convolution neural networks- (CNN-) based frameworks on several ReID benchmarks. Furthermore, two modules are designed in consideration of the specialties of ReID data: (1) It is super natural and simple for Transformer to encode non-visual information such as camera or viewpoint into vector embedding representations. Plugging into these embeddings, ViT holds the ability to eliminate the bias caused by diverse cameras or viewpoints.(2) We design a Jigsaw branch, parallel with the Global branch, to facilitate the training of the model in a two-branch learning framework. In the Jigsaw branch, a jigsaw patch module is designed to learn robust feature representation and help the training of transformer by shuffling the patches. With these novel modules, we propose a pure-transformer framework dubbed as TransReID, which is the first work to use a pure Transformer for ReID research to the best of our knowledge. Experimental results of TransReID are superior promising, which achieve state-of-the-art performance on both person and vehicle ReID benchmarks.
Vision-and-Language Pretraining (VLP) has improved performance on various joint vision-and-language downstream tasks. Current approaches for VLP heavily rely on image feature extraction processes, most of which involve region supervisions (e.g., object detection) and the convolutional architecture (e.g., ResNet). Although disregarded in the literature, we find it problematic in terms of both (1) efficiency/speed, that simply extracting input features requires much more computation than the actual multimodal interaction steps; and (2) expressive power, as it is upper bounded to the expressive power of the visual encoder and its predefined visual vocabulary. In this paper, we present a minimal VLP model, Vision-and-Language Transformer (ViLT), monolithic in the sense that processing of visual inputs is drastically simplified to just the same convolution-free manner that we process textual inputs. We show that ViLT is up to 60 times faster than previous VLP models, yet with competitive or better downstream task performance.
We present BoTNet, a conceptually simple yet powerful backbone architecture that incorporates self-attention for multiple computer vision tasks including image classification, object detection and instance segmentation. By just replacing the spatial convolutions with global self-attention in the final three bottleneck blocks of a ResNet and no other changes, our approach improves upon the baselines significantly on instance segmentation and object detection while also reducing the parameters, with minimal overhead in latency. Through the design of BoTNet, we also point out how ResNet bottleneck blocks with self-attention can be viewed as Transformer blocks. Without any bells and whistles, BoTNet achieves 44.4% Mask AP and 49.7% Box AP on the COCO Instance Segmentation benchmark using the Mask R-CNN framework; surpassing the previous best published single model and single scale results of ResNeSt evaluated on the COCO validation set. Finally, we present a simple adaptation of the BoTNet design for image classification, resulting in models that achieve a strong performance of 84.7% top-1 accuracy on the ImageNet benchmark while being up to 2.33x faster in compute time than the popular EfficientNet models on TPU-v3 hardware. We hope our simple and effective approach will serve as a strong baseline for future research in self-attention models for vision.
This paper presents VTN, a transformer-based framework for video recognition. Inspired by recent developments in vision transformers, we ditch the standard approach in video action recognition that relies on 3D ConvNets and introduce a method that classifies actions by attending to the entire video sequence information. Our approach is generic and builds on top of any given 2D spatial network. In terms of wall runtime, it trains 16.1× faster and runs 5.1× faster during inference while maintaining competitive accuracy compared to other state-of-the-art methods. It enables whole video analysis, via a single end-to-end pass, while requiring 1.5× fewer GFLOPs. We report competitive results on Kinetics-400 and present an ablation study of VTN properties and the trade-off between accuracy and inference speed. We hope our approach will serve as a new baseline and start a fresh line of research in the video recognition domain.
In this paper we introduce a Transformer-based approach to video object segmentation (VOS). To address compounding error and scalability issues of prior work, we propose a scalable, end-to-end method for VOS called Sparse Spatiotemporal Transformers (SST). SST extracts per-pixel representations for each object in a video using sparse attention over spatiotemporal features. Our attention-based formulation for VOS allows a model to learn to attend over a history of multiple frames and provides suitable inductive bias for performing correspondence-like computations necessary for solving motion segmentation. We demonstrate the effectiveness of attention-based over recurrent networks in the spatiotemporal domain. Our method achieves competitive results on YouTube-VOS and DAVIS 2017 with improved scalability and robustness to occlusions compared with the state of the art.
This work presents a new fine-grained transparent object segmentation dataset, termed Trans10K-v2, extending Trans10K-v1, the first large-scale transparent object segmentation dataset. Unlike Trans10K-v1 that only has two limited categories, our new dataset has several appealing benefits. (1) It has 11 fine-grained categories of transparent objects, commonly occurring in the human domestic environment, making it more practical for real-world application. (2) Trans10K-v2 brings more challenges for the current advanced segmentation methods than its former version. Furthermore, a novel transformer-based segmentation pipeline termed Trans2Seg is proposed. Firstly, the transformer encoder of Trans2Seg provides the global receptive field in contrast to CNN’s local receptive field, which shows excellent advantages over pure CNN architectures. Secondly, by formulating semantic segmentation as a problem of dictionary look-up, we design a set of learnable prototypes as the query of Trans2Seg’s transformer decoder, where each prototype learns the statistics of one category in the whole dataset. We benchmark more than 20 recent semantic segmentation methods, demonstrating that Trans2Seg significantly outperforms all the CNN-based methods, showing the proposed algorithm’s potential ability to solve transparent object segmentation.
Text-to-image retrieval is an essential task in multi-modal information retrieval, i.e. retrieving relevant images from a large and unlabelled image dataset given textual queries. In this paper, we propose VisualSparta, a novel text-to-image retrieval model that shows substantial improvement over existing models on both accuracy and efficiency. We show that VisualSparta is capable of outperforming all previous scalable methods in MSCOCO and Flickr30K. It also shows substantial retrieving speed advantages, i.e. for an index with 1 million images, VisualSparta gets over 391x speed up compared to standard vector search. Experiments show that this speed advantage even gets bigger for larger datasets because VisualSparta can be efficiently implemented as an inverted index. To the best of our knowledge, VisualSparta is the first transformer-based text-to-image retrieval model that can achieve real-time searching for very large dataset, with significant accuracy improvement compared to previous state-of-the-art methods.
Most existing point-cloud based 3D object detectors use convolution-like operators to process information in a local neighbourhood with fixed-weight kernels and aggregate global context hierarchically. However, recent work on non-local neural networks and self-attention for 2D vision has shown that explicitly modeling global context and long-range interactions between positions can lead to more robust and competitive models. In this paper, we explore two variants of self-attention for contextual modeling in 3D object detection by augmenting convolutional features with self-attention features. We first incorporate the pairwise self-attention mechanism into the current state-of-the-art BEV, voxel and point-based detectors and show consistent improvement over strong baseline models while simultaneously significantly reducing their parameter footprint and computational cost. We also propose a self-attention variant that samples a subset of the most representative features by learning deformations over randomly sampled locations. This not only allows us to scale explicit global contextual modeling to larger point-clouds, but also leads to more discriminative and informative feature descriptors. Our method can be flexibly applied to most state-of-the-art detectors with increased accuracy and parameter and compute efficiency. We achieve new state-of-the-art detection performance on KITTI and nuScenes datasets.