CVJun 25, 2021

Image-to-image Transformation with Auxiliary Condition

arXiv:2106.13696v1
Originality Incremental advance
AI Analysis

This addresses the issue of domain adaptation in image recognition for applications like surveillance, but it is incremental as it builds on existing CycleGAN methods.

The paper tackles the problem of GAN-based image-to-image transformation methods being insensitive to variations in pose and shape when training data is imbalanced, by introducing label information into CycleGAN to create label-wise transformation models. The result is evaluated on digit image transformation from SVHN to MNIST and surveillance camera image transformation from simulated to real images, though no concrete numbers are provided.

The performance of image recognition like human pose detection, trained with simulated images would usually get worse due to the divergence between real and simulated data. To make the distribution of a simulated image close to that of real one, there are several works applying GAN-based image-to-image transformation methods, e.g., SimGAN and CycleGAN. However, these methods would not be sensitive enough to the various change in pose and shape of subjects, especially when the training data are imbalanced, e.g., some particular poses and shapes are minor in the training data. To overcome this problem, we propose to introduce the label information of subjects, e.g., pose and type of objects in the training of CycleGAN, and lead it to obtain label-wise transforamtion models. We evaluate our proposed method called Label-CycleGAN, through experiments on the digit image transformation from SVHN to MNIST and the surveillance camera image transformation from simulated to real images.

Foundations

The foundational work for this paper's niche, ranked by how specifically the neighbourhood builds on it — not by global fame.

Your Notes