Yixin Jin
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Yixin Jin has a total of 5 co-authors. Here are the top co-authors by number of shared publications:
- Honglak Lee
University of Michigan
1 shared publication - Yijie Guo
University of Michigan
1 shared publication Yijun Luo
University of Michigan Ann Arbor
1 shared publication- Yuting Zhang
University of Michigan Ann Arbor MI USA
1 shared publication - Zhiyuan He
University of Michigan Ann Arbor MI USA
1 shared publication
Latest Publications
- Open Access
Unsupervised Discovery of Object Landmarks as Structural Representations
Deep neural networks can model images with rich latent representations, but they cannot naturally conceptualize structures of object categories in a human-perceptible way. This paper addresses the problem of learning object structures in an image modeling process without supervision. We propose an autoencoding formulation to discover landmarks as explicit structural representations. The encoding module outputs landmark coordinates, whose validity is ensured by constraints that reflect the necessary properties for landmarks. The decoding module takes the landmarks as a part of the learnable input representations in an end-to-end differentiable framework. Our discovered landmarks are semantically meaningful and more predictive of manually annotated landmarks than those discovered by previous methods. The coordinates of our landmarks are also complementary features to pretrained deep-neural-network representations in recognizing visual attributes. In addition, the proposed method naturally creates an unsupervised, perceptible interface to manipulate object shapes and decode images with controllable structures. The project webpage is at http://ytzhang.net/projects/lmdis-rep
Open AccessUnsupervised Discovery of Object Landmarks as Structural Representations
Deep neural networks can model images with rich latent representations, but they cannot naturally conceptualize structures of object categories in a human-perceptible way. This paper addresses the problem of learning object structures in an image modeling process without supervision. We propose an autoencoding formulation to discover landmarks as explicit structural representations. The encoding module outputs landmark coordinates, whose validity is ensured by constraints that reflect the necessary properties for landmarks. The decoding module takes the landmarks as a part of the learnable input representations in an end-to-end differentiable framework. Our discovered landmarks are semantically meaningful and more predictive of manually annotated landmarks than those discovered by previous methods. The coordinates of our landmarks are also complementary features to pretrained deep-neural-network representations in recognizing visual attributes. In addition, the proposed method naturally creates an unsupervised, perceptible interface to manipulate object shapes and decode images with controllable structures. The project webpage is at http://ytzhang.net/projects/lmdis-rep