Vision Transformers are deep learning models that apply the transformer architecture to visual data by representing images as sequences of patches and processing them with self-attention mechanisms. Unlike convolutional neural networks, they capture long-range dependencies and global context directly through attention. Vision transformers are used in tasks such as image classification, object detection, and segmentation, often achieving state-of-the-art performance when trained on large datasets.

Posts

Efficient Transformer Encoders for Mask2Former-style Models

/in /by

Vision transformer based models bring significant improvements for image segmentation tasks. Although these architectures offer powerful capabilities irrespective of specific segmentation tasks, their use of computational resources can be taxing on deployed devices. One way to overcome this challenge is by adapting the computation level to the specific needs of the input image rather than the current one size-fits-all approach. To this end, we introduce ECO-M2F or EffiCient TransfOrmer Encoders for Mask2Former-style models. Noting that the encoder module of M2F-style models incur high resource-intensive computations, ECO-M2F provides a strategy to self-select the number of hidden layers in the encoder, conditioned on the input image. To enable this self-selection ability for providing a balance between performance and computational efficiency, we present a three-step recipe. The first step is to train the parent architecture to enable early exiting from the encoder. The second step is to create a derived dataset of the ideal number of encoder layers required for each training example. The third step is to use the aforementioned derived dataset to train a gating network that predicts the number of encoder layers to be used, conditioned on input image. Additionally, to change the computational-accuracy trade off, only steps two and three need to be repeated which significantly reduces retraining time. Experiments on the public datasets show that the proposed approach reduces expected encoder computational cost while maintaining performance, adapts to various user compute resources, is flexible in architecture configurations, and can be extended beyond the segmentation task to object detection.

Read more

FactionFormer: Context-Driven Collaborative Vision Transformer Models for Edge Intelligence

/in /by

Edge Intelligence has received attention in the recent times for its potential towards improving responsiveness, reducing the cost of data transmission, enhancing security and privacy, and enabling autonomous decisions by edge devices. However, edge devices lack the power and compute resources necessary to execute most Al models. In this paper, we present FactionFormer, a novel method to deploy resource-intensive deep-learning models, such as vision transformers (ViT), on resource-constrained edge devices. Our method is based on a key observation: edge devices are often deployed in settings where they encounter only a subset of the classes that the resource intensive Al model is trained to classify, and this subset changes across deployments. Therefore, we automatically identify this subset as a faction, devise on-the fly a bespoke resource-efficient ViT called a modelette for the faction and set up an efficient processing pipeline consisting of a modelette on the device, a wireless network such as 5G, and the resource-intensive ViT model on an edge server, all of which work collaboratively to do the inference. For several ViT models pre-trained on benchmark datasets, FactionFormer’s modelettes are up to 4× smaller than the corresponding baseline models in terms of the number of parameters, and they can infer up to 2.5× faster than the baseline setup where every input is processed by the resource-intensive ViT on the edge server. Our work is the first of its kind to propose a device-edge collaborative inference framework where bespoke deep learning models for the device are automatically devised on-the-fly for most frequently encountered subset of classes.

Read more