We address the problem of generating human motion sequences that can be conditioned on three context categories: semantic motion information, eg target action/object labels, geometric scene information and past or future motion observations. State-of-the-art methods consist of either models specialized to one single setting or unconditional models which have by nature limited applicability. In contrast, we introduce \ours, a generalized approach that can generate human motion sequences of variable length conditioned with various combinations of the three contextual categories. It can leverage large amounts of unconditional data, and be used in various settings depending on the desired control and available information. $\mathrm{Purposer}$ addresses the problem in a two-stage fashion, inspired by neural discrete representation learning. First, it encodes unconditional human motion into a discrete latent space. Second, a generative model trained for next-step prediction in this space synthesizes sequences of latent indices. Our model can be conditioned on any combination of available context. In contrast to most existing methods, ours is learning-based and does not rely on a test time optimization loop. Our experimental results show that besides offering more controllability than existing methods, our model generates motion sequences that are compelling, diverse, and coherent with the given contextual information.
Purposer: putting human motion generation in context
| Nicolás Ugrinovic, Thomas Lucas, Fabien Baradel, Philippe Weinzaepfel, Gregory Rogez, Francesc Moreno-Noguer |
| International Conference on 3D Vision (3DV), Davos, Switzerland, 18-21 March, 2024 |
To make robots autonomous in real-world everyday spaces, they should be able to learn from their interactions within these spaces, how to best execute tasks specified by non-expert users in a safe and reliable way. To do so requires sequential decision-making skills that combine machine learning, adaptive planning and control in uncertain environments as well as solving hard combinatorial optimisation problems. Our research combines expertise in reinforcement learning, computer vision, robotic control, sim2real transfer, large multimodal foundation models and neural combinatorial optimisation to build AI-based architectures and algorithms to improve robot autonomy and robustness when completing everyday complex tasks in constantly changing environments.
The research we conduct on expressive visual representations is applicable to visual search, object detection, image classification and the automatic extraction of 3D human poses and shapes that can be used for human behavior understanding and prediction, human-robot interaction or even avatar animation. We also extract 3D information from images that can be used for intelligent robot navigation, augmented reality and the 3D reconstruction of objects, buildings or even entire cities.
Our work covers the spectrum from unsupervised to supervised approaches, and from very deep architectures to very compact ones. We’re excited about the promise of big data to bring big performance gains to our algorithms but also passionate about the challenge of working in data-scarce and low-power scenarios.
Furthermore, we believe that a modern computer vision system needs to be able to continuously adapt itself to its environment and to improve itself via lifelong learning. Our driving goal is to use our research to deliver embodied intelligence to our users in robotics, autonomous driving, via phone cameras and any other visual means to reach people wherever they may be.
This web site uses cookies for the site search, to display videos and for aggregate site analytics.
Learn more about these cookies in our privacy notice.
Cookie settings
You may choose which kind of cookies you allow when visiting this website. Click on "Save cookie settings" to apply your choice.
This content is currently blocked. To view the content please either 'Accept social media cookies' or 'Accept all cookies'.
For more information on cookies see our privacy notice.