# Estimating Motion Codes from Demonstration Videos

arXiv: https://arxiv.org/abs/2007.15841 · HTML: https://mlsanity.com/paper/2007.15841/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 31 Jul 2020.

## Built on by (most cited first, 1 of 1)

- [Calorie Aware Automatic Meal Kit Generation from an Image](https://mlsanity.com/paper/2112.09839/) (2021)

## Abstract (arXiv)

A motion taxonomy can encode manipulations as a binary-encoded representation, which we refer to as motion codes. These motion codes innately represent a manipulation action in an embedded space that describes the motion's mechanical features, including contact and trajectory type. The key advantage of using motion codes for embedding is that motions can be more appropriately defined with robotic-relevant features, and their distances can be more reasonably measured using these motion features. In this paper, we develop a deep learning pipeline to extract motion codes from demonstration videos in an unsupervised manner so that knowledge from these videos can be properly represented and used for robots. Our evaluations show that motion codes can be extracted from demonstrations of action in the EPIC-KITCHENS dataset.