CR-GAN: Learning Complete Representations for Multi-view Generation

Authors: Yu Tian, Xi Peng, Long Zhao, Shaoting Zhang, Dimitris N. Metaxas

IJCAI 2018 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental The experimental results prove that CR-GAN significantly outperforms stateof-the-art methods, especially when generating from unseen inputs in wild conditions. 4 Experiments CR-GAN aims to learn complete representations in the embedding space. We achieve this goal by combining the two-pathway architecture with self-supervised learning. We conduct experiments to evaluate these two contributions respectively. Then we compare our CR-GAN with DR-GAN [Tran et al., 2017], both the visual results and t-SNE visualization in the embedding space are shown. We also compare CR-GAN and Bi GAN with an image reconstruction task.
Researcher Affiliation Academia Yu Tian1, Xi Peng1, Long Zhao1, Shaoting Zhang2 and Dimitris N. Metaxas1 1 Rutgers University 2 University of North Carolina at Charlotte {yt219, px13, lz311, dnm}@cs.rutgers.edu, szhang16@uncc.edu
Pseudocode Yes Algorithm 1: Supervised training with two paths
Open Source Code Yes 1 The code and pre-trained models are publicly available: https://github.com/bluer555/CR-GAN
Open Datasets Yes Multi-PIE [Gross et al., 2010] is a labeled dataset collected under constrained environment. We use 250 subjects from the first session with 9 poses within 60 , 20 illuminations, and two expressions. The first 200 subjects are for training and the rest 50 for testing. 300w LP [Zhu et al., 2016] is augmented from 300W [Sagonas et al., 2013] by the face profiling approach [Zhu et al., 2016], which contains view labels as well. We employ images with yaw angles ranging from 60 to +60 , and discretize them into 9 intervals. For evaluation on unlabeled datasets, we use Celeb A [Liu et al., 2015] and IJB-A [Klare et al., 2015].
Dataset Splits No The paper states, 'The first 200 subjects are for training and the rest 50 for testing' for Multi-PIE, but does not explicitly provide details about a validation dataset split.
Hardware Specification No The paper does not provide specific details on the hardware used, such as GPU or CPU models.
Software Dependencies No The paper mentions using WGAN-GP and Adam optimizer but does not specify version numbers for any software libraries or dependencies.
Experiment Setup Yes During training, we set v to be a one-hot vector with 9 dimensions and z [ 1, 1]119 in the latent space. The batch size is 64. Adam optimizer [Kingma and Ba, 2015] is used with the learning rate of 0.0005 and momentum of [0, 0.9]. According to the setting of WGAN-GP, we let λ1 = 10, λ2 λ4 = 1, λ5 = 0.01. Moreover, all the networks are trained after 25 epochs in supervised learning; we train 10 more epochs in self-supervised learning.