# Multiview Supervision By Registration

arXiv: https://arxiv.org/abs/1811.11251 · HTML: https://mlsanity.com/paper/1811.11251/

## Key facts

- Cited by 7 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 27 Nov 2018.

## Built on by (most cited first, 7 of 7)

- [Supervision by Registration and Triangulation for Landmark Detection](https://mlsanity.com/paper/2101.09866/) (2021)
- [3D Human Reconstruction in the Wild with Collaborative Aerial Cameras](https://mlsanity.com/paper/2108.03936/) (2021)
- [SemiMultiPose: A Semi-supervised Multi-animal Pose Estimation Framework](https://mlsanity.com/paper/2204.07072/) (2022)
- [A Horse with no Labels: Self-Supervised Horse Pose Estimation from Unlabelled Images and Synthetic Prior](https://mlsanity.com/paper/2308.03411/) (2023)
- [Towards Multi-Modal Animal Pose Estimation: A Survey and In-Depth Analysis](https://mlsanity.com/paper/2410.09312/) (2024)
- [Pose Splatter: A 3D Gaussian Splatting Model for Quantifying Animal Pose and Appearance](https://mlsanity.com/paper/2505.18342/) (2025)
- [MBW: Multi-view Bootstrapping in the Wild](https://mlsanity.com/paper/2210.01721/) (2022)

## Abstract (arXiv)

This paper presents a semi-supervised learning framework to train a keypoint detector using multiview image streams given the limited labeled data (typically <4%). We leverage the complementary relationship between multiview geometry and visual tracking to provide three types of supervisionary signals to utilize the unlabeled data: (1) keypoint detection in one view can be supervised by other views via the epipolar geometry; (2) a keypoint moves smoothly over time where its optical flow can be used to temporally supervise consecutive image frames to each other; (3) visible keypoint in one view is likely to be visible in the adjacent view. We integrate these three signals in a differentiable fashion to design a new end-to-end neural network composed of three pathways. This design allows us to extensively use the unlabeled data to train the keypoint detector. We show that our approach outperforms existing detectors including DeepLabCut tailored to the keypoint detection of non-human species such as monkeys, dogs, and mice.