# Unsupervised learning of object frames by dense equivariant image labelling

arXiv: https://arxiv.org/abs/1706.02932 · HTML: https://mlsanity.com/paper/1706.02932/

## Key facts

- Cited by 8 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 9 Jun 2017.

## Built on by (most cited first, 8 of 8)

- [Unsupervised Object-Level Representation Learning from Scene Images](https://mlsanity.com/paper/2106.11952/) (2021)
- [GAN-Supervised Dense Visual Alignment](https://mlsanity.com/paper/2112.05143/) (2021)
- [Template NeRF: Towards Modeling Dense Shape Correspondences from Category-Specific Object Images](https://mlsanity.com/paper/2111.04237/) (2021)
- [TUSK: Task-Agnostic Unsupervised Keypoints](https://mlsanity.com/paper/2206.08460/) (2022)
- [Landmarks Augmentation with Manifold-Barycentric Oversampling](https://mlsanity.com/paper/2104.00925/) (2021)
- [Pretrained equivariant features improve unsupervised landmark discovery](https://mlsanity.com/paper/2104.02925/) (2021)
- [Pose Recognition in the Wild: Animal pose estimation using Agglomerative Clustering and Contrastive Learning](https://mlsanity.com/paper/2111.08259/) (2021)
- [DenseSeg: Joint Learning for Semantic Segmentation and Landmark Detection Using Dense Image-to-Shape Representation](https://mlsanity.com/paper/2405.19746/) (2024)

## Abstract (arXiv)

One of the key challenges of visual perception is to extract abstract models of 3D objects and object categories from visual measurements, which are affected by complex nuisance factors such as viewpoint, occlusion, motion, and deformations. Starting from the recent idea of viewpoint factorization, we propose a new approach that, given a large number of images of an object and no other supervision, can extract a dense object-centric coordinate frame. This coordinate frame is invariant to deformations of the images and comes with a dense equivariant labelling neural network that can map image pixels to their corresponding object coordinates. We demonstrate the applicability of this method to simple articulated objects and deformable objects such as human faces, learning embeddings from random synthetic transformations or optical flow correspondences, all without any manual supervision.