A Study on GANs based on Pose Manifold for Rigid Object Pose Estimation

Generative Adversarial Nets (GANs) is a pair of neural networks which can learn data distribution and generate various data from the distribution. In this research, by focusing on the fact that pose variation of a rigid object can be expressed on a manifold in a latent space, we introduce a GANs model which generates data from a distribution defined over a manifold. We also propose a pose estimation method which trains a pose estimator while interpolating training images using the GANs. We evaluated the interpolation capability of the proposed model using a public dataset, and also evaluated pose estimation accuracy of the proposed model.

[1]  Tomas Pfister,et al.  Learning from Simulated and Unsupervised Images through Adversarial Training , 2016, 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[2]  Jitendra Malik,et al.  Aligning 3D models to RGB-D images of cluttered scenes , 2015, 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[3]  Yoshua Bengio,et al.  Generative Adversarial Nets , 2014, NIPS.

[4]  Iasonas Kokkinos,et al.  Deep Filter Banks for Texture Recognition, Description, and Segmentation , 2015, International Journal of Computer Vision.

[5]  Alexei A. Efros,et al.  Image-to-Image Translation with Conditional Adversarial Networks , 2016, 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[6]  Geoffrey E. Hinton,et al.  ImageNet classification with deep convolutional neural networks , 2012, Commun. ACM.

[7]  Simon Osindero,et al.  Conditional Generative Adversarial Nets , 2014, ArXiv.

[8]  Soumith Chintala,et al.  Unsupervised Representation Learning with Deep Convolutional Generative Adversarial Networks , 2015, ICLR.

[9]  Sameer A. Nene,et al.  Columbia Object Image Library (COIL100) , 1996 .

[10]  Luc Van Gool,et al.  Crossing Nets: Combining GANs and VAEs with a Shared Latent Space for Hand Pose Estimation , 2017, 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[11]  Hiroshi Murase,et al.  Visual learning and recognition of 3-d objects from appearance , 2005, International Journal of Computer Vision.

[12]  J. Broekens,et al.  Assistive social robots in elderly care: a review , 2009 .

[13]  Ahmed M. Elgammal,et al.  Regression from local features for viewpoint and pose estimation , 2011, 2011 International Conference on Computer Vision.