Skip to search formSkip to main contentSkip to account menu

Implicit Semantic Data Augmentation for Deep Networks

This work shows that the proposed ISDA amounts to minimizing a novel robust CE loss, which adds negligible extra computational cost to a normal training procedure, and consistently improves the generalization performance of popular deep models (ResNets and DenseNets) on a variety of datasets.
View PDF on arXiv (opens in a new tab)

Regularizing Deep Networks With Semantic Data Augmentation

The proposed implicit semantic data augmentation (ISDA) algorithm amounts to minimizing a novel robust CE loss, which adds minimal extra computational cost to a normal training procedure, and can be applied to semi-supervised learning tasks under the consistency regularization framework.
View on IEEE (opens in a new tab)

Adaptive Focus for Efficient Video Recognition

This paper model the patch localization problem as a sequential decision task, and proposes a reinforcement learning based approach for efficient spatially adaptive video recognition (AdaFocus), whose features are used by a recurrent policy network to localize the most task-relevant regions.
View on IEEE (opens in a new tab)

Revisiting Locally Supervised Learning: an Alternative to End-to-end Training

An information propagation (InfoPro) loss is proposed, which encourages local modules to preserve as much useful information as possible, while progressively discard task-irrelevant information, and is capable of achieving competitive performance with less than 40% memory footprint compared to E2E training.
View PDF on arXiv (opens in a new tab)

Dynamic Neural Networks: A Survey

This survey comprehensively review this rapidly developing area of dynamic networks by dividing dynamic networks into three main categories: sample-wise dynamic models that process each sample with data-dependent architectures or parameters; spatial-wiseynamic networks that conduct adaptive computation with respect to different spatial locations of image data; and temporal-wise Dynamic networks that perform adaptive inference along the temporal dimension for sequential data.
View on IEEE (opens in a new tab)

Glance and Focus: a Dynamic Approach to Reducing Spatial Redundancy in Image Classification

This work proposes a novel framework that performs efficient image classification by processing a sequence of relatively small inputs, which are strategically selected from the original image with reinforcement learning, which consistently improves the computational efficiency of a wide variety of deep models.
View PDF on arXiv (opens in a new tab)

MetaSAug: Meta Semantic Augmentation for Long-Tailed Visual Recognition

This paper addresses the issue of imbalance in real-world training data by augmenting minority classes with a recently proposed implicit semantic data augmentation (ISDA) algorithm, which produces diversified augmented samples by translating deep features along many semantically meaningful directions.
View on IEEE (opens in a new tab)

Not All Images are Worth 16x16 Words: Dynamic Vision Transformers with Adaptive Sequence Length

This paper argues that every image has its own characteristics, and ideally the token number should be conditioned on each individual input, and proposes a Dynamic Transformer to automatically configure a proper number of tokens for each input image.

Not All Images are Worth 16x16 Words: Dynamic Transformers for Efficient Image Recognition

This paper argues that every image has its own characteristics, and ideally the token number should be conditioned on each individual input, and proposes a Dynamic Transformer to automatically configure a proper number of tokens for each input image.
View PDF on arXiv (opens in a new tab)

Transferable Semantic Augmentation for Domain Adaptation

This work proposes a Transferable Semantic Augmentation approach to enhance the classifier adaptation ability through implicitly generating source features towards target semantics, inspired by the fact that deep feature transformation towards a certain direction can be represented as meaningful semantic altering in the original input space.
View on IEEE (opens in a new tab)
By clicking accept or continuing to use the site, you agree to the terms outlined in our Privacy Policy (opens in a new tab), Terms of Service (opens in a new tab), and Dataset License (opens in a new tab)