Unsupervised Discovery of Object Landmarks as Structural Representations

04/12/2018
by   Yuting Zhang, et al.
0

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

READ FULL TEXT

page 21

page 29

page 31

page 33

page 34

page 35

page 36

page 37

research
05/05/2017

Unsupervised learning of object landmarks by factorized spatial embeddings

Learning automatically the structure of object categories remains an imp...
research
06/26/2020

Unsupervised Discovery of Object Landmarks via Contrastive Learning

Given a collection of images, humans are able to discover landmarks of t...
research
09/19/2023

Unsupervised Landmark Discovery Using Consistency Guided Bottleneck

We study a challenging problem of unsupervised discovery of object landm...
research
04/22/2021

KeypointDeformer: Unsupervised 3D Keypoint Discovery for Shape Control

We introduce KeypointDeformer, a novel unsupervised method for shape con...
research
06/30/2015

Discovering Characteristic Landmarks on Ancient Coins using Convolutional Networks

In this paper, we propose a novel method to find characteristic landmark...
research
06/29/2020

Unsupervised Landmark Learning from Unpaired Data

Recent attempts for unsupervised landmark learning leverage synthesized ...
research
06/28/2023

Identifiability of Discretized Latent Coordinate Systems via Density Landmarks Detection

Disentanglement aims to recover meaningful latent ground-truth factors f...

Please sign up or login with your details

Forgot password? Click here to reset