AGA: Attribute-Guided Augmentation

Dixit, Mandar; Kwitt, Roland; Niethammer, Marc; Vasconcelos, Nuno

doi:10.1109/cvpr.2017.355

Cited by 98 publications

(72 citation statements)

References 23 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…III-A. Semantic attributes can be pre-defined by human experts [14]. Semantic word vector u base zi is the projection of each vocabulary entity Encoder TriNet is composed of four layers corresponding to each layer of ResNet-18.…”

Section: Dual Trinet Networkmentioning

confidence: 99%

See 1 more Smart Citation

Multi-Level Semantic Feature Augmentation for One-Shot Learning

Chen

Zhang

et al. 2019

IEEE Trans. on Image Process.

229

106

View full text Add to dashboard Cite

The ability to quickly recognize and learn new visual concepts from limited samples enables humans to quickly adapt to new tasks and environments. This ability is enabled by semantic association of novel concepts with those that have already been learned and stored in memory. Computers can start to ascertain similar abilities by utilizing a semantic concept space. A concept space is a high-dimensional semantic space in which similar abstract concepts appear close and dissimilar ones far apart. In this paper, we propose a novel approach to one-shot learning that builds on this core idea. Our approach learns to map a novel sample instance to a concept, relates that concept to the existing ones in the concept space and, using these relationships, generates new instances, by interpolating among the concepts, to help learning. Instead of synthesizing new image instance, we propose to directly synthesize instance features by leveraging semantics using a novel auto-encoder network we call dual TriNet. The encoder part of the TriNet learns to map multi-layer visual features from CNN to a semantic vector. In semantic space, we search for related concepts, which are then projected back into the image feature spaces by the decoder portion of the TriNet. Two strategies in the semantic space are explored. Notably, this seemingly simple strategy results in complex augmented feature distributions in the image feature space, leading to substantially better performance.

show abstract

Section: Dual Trinet Networkmentioning

confidence: 99%

“…HOG) of the object parts and combined them to synthesize new feature representations. Dixit et al [14], for the first time, considered attributesguided augmentation to synthesize sample features. Their work, however, utilizes and relies on a set of pre-defined semantic attributes.…”

Section: Introductionmentioning

confidence: 99%

Multi-Level Semantic Feature Augmentation for One-Shot Learning

Chen

Zhang

et al. 2019

IEEE Trans. on Image Process.

229

106

View full text Add to dashboard Cite

show abstract

“…For the same purpose (i.e. data augmentation for data-starved classes), in [26] the authors propose an attributed-guided augmentation approach which learns a mapping that allows the creation of synthetic data by manipulating certain attributes of real data. Thus, the newly created data presents attributes based on user-defined criteria (values).…”

Section: Zagoruyko and Komodakismentioning

confidence: 99%

Saliency for fine-grained object recognition in domains with scarce training data

Flores

González-García

Weijer

et al. 2019

Pattern Recognition

View full text Add to dashboard Cite

This paper investigates the role of saliency to improve the classification accuracy of a Convolutional Neural Network (CNN) for the case when scarce training data is available. Our approach consists in adding a saliency branch to an existing CNN architecture which is used to modulate the standard bottom-up visual features from the original image input, acting as an attentional mechanism that guides the feature extraction process. The main aim of the proposed approach is to enable the effective training of a fine-grained recognition model with limited training samples and to improve the performance on the task, thereby alleviat-* to saliency methods that yield improved performance gains when applied in an object recognition pipeline.

show abstract

“…Implementation of this augmentation in the real world, especially for long audio sequences of high dimension is not optimal. One potential solution could be multivariate distribution learning in representation space [43] with respect to the structural components [44] of a spectrogram.…”

Section: Introductionmentioning

confidence: 99%

Unsupervised feature learning for environmental sound classification using Weighted Cycle-Consistent Generative Adversarial Network

Esmaeilpour

Cardinal

Koerich

2020

Applied Soft Computing

View full text Add to dashboard Cite

In this paper we propose a novel environmental sound classification approach incorporating unsupervised feature learning via the spherical K-Means++ algorithm and a new architecture for high-level data augmentation. The audio signal is transformed into a 2D representation using a discrete wavelet transform (DWT). The DWT spectrograms are then augmented by a novel architecture for cycle-consistent generative adversarial network. This high-level augmentation bootstraps generated spectrograms in both intra-and inter-class manners by translating structural features from sample to sample. A codebook is built by coding the DWT spectrograms with the speeded-up robust feature detector and the K-Means++ algorithm. The Random forest is the final learning algorithm which learns the environmental sound classification task from the code vectors. Experimental results in four benchmarking environmental sound datasets (ESC-10, ESC-50, UrbanSound8k, and DCASE-2017) have shown that the proposed classification approach outperforms most of the state-of-the-art classifiers, including convolutional neural networks such as AlexNet and GoogLeNet, improving the classification rate between 3.51% and 14.34%, depending on the dataset.

show abstract

AGA: Attribute-Guided Augmentation

Cited by 98 publications

References 23 publications

Multi-Level Semantic Feature Augmentation for One-Shot Learning

Multi-Level Semantic Feature Augmentation for One-Shot Learning

Saliency for fine-grained object recognition in domains with scarce training data

Unsupervised feature learning for environmental sound classification using Weighted Cycle-Consistent Generative Adversarial Network

Contact Info

Product

Resources

About