DiffMimic: Efficient Motion Mimicking with Differentiable Physics
- URL: http://arxiv.org/abs/2304.03274v2
- Date: Wed, 26 Apr 2023 06:29:21 GMT
- Title: DiffMimic: Efficient Motion Mimicking with Differentiable Physics
- Authors: Jiawei Ren, Cunjun Yu, Siwei Chen, Xiao Ma, Liang Pan, Ziwei Liu
- Abstract summary: We leverage differentiable physics simulators (DPS) and propose an efficient motion mimicking method dubbed DiffMimic.
Our key insight is that DPS casts a complex policy learning task to a much simpler state matching problem.
Extensive experiments on standard benchmarks show that DiffMimic has a better sample efficiency and time efficiency than existing methods.
- Score: 41.442225872857136
- License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
- Abstract: Motion mimicking is a foundational task in physics-based character animation.
However, most existing motion mimicking methods are built upon reinforcement
learning (RL) and suffer from heavy reward engineering, high variance, and slow
convergence with hard explorations. Specifically, they usually take tens of
hours or even days of training to mimic a simple motion sequence, resulting in
poor scalability. In this work, we leverage differentiable physics simulators
(DPS) and propose an efficient motion mimicking method dubbed DiffMimic. Our
key insight is that DPS casts a complex policy learning task to a much simpler
state matching problem. In particular, DPS learns a stable policy by analytical
gradients with ground-truth physical priors hence leading to significantly
faster and stabler convergence than RL-based methods. Moreover, to escape from
local optima, we utilize a Demonstration Replay mechanism to enable stable
gradient backpropagation in a long horizon. Extensive experiments on standard
benchmarks show that DiffMimic has a better sample efficiency and time
efficiency than existing methods (e.g., DeepMimic). Notably, DiffMimic allows a
physically simulated character to learn Backflip after 10 minutes of training
and be able to cycle it after 3 hours of training, while the existing approach
may require about a day of training to cycle Backflip. More importantly, we
hope DiffMimic can benefit more differentiable animation systems with
techniques like differentiable clothes simulation in future research.
Related papers
- Improving Gradient Computation for Differentiable Physics Simulation
with Contacts [10.450509067356148]
We study differentiable rigid-body simulation with contacts.
We propose to improve gradient computation by continuous collision detection and leverage the time-of-impact (TOI)
We show that with TOI-Ve, we are able to learn an optimal control sequence that matches the analytical solution.
arXiv Detail & Related papers (2023-04-28T21:10:16Z) - MoQuad: Motion-focused Quadruple Construction for Video Contrastive
Learning [10.41936704731324]
This paper presents a simple yet effective sample construction strategy to boost the learning of motion features in video contrastive learning.
The proposed method, dubbed Motion-focused Quadruple Construction (MoQuad), augments the instance discrimination by meticulously disturbing the appearance and motion of both the positive and negative samples.
By simply applying MoQuad to SimCLR, extensive experiments show that we achieve superior performance on downstream tasks compared to the state of the arts.
arXiv Detail & Related papers (2022-12-21T09:26:40Z) - Imitation Learning via Differentiable Physics [26.356669151969953]
Imitation learning (IL) methods such as inverse reinforcement learning (IRL) usually have a double-loop training process.
We propose a new IL method, i.e., Imitation Learning via Differentiable Physics (ILD), which gets rid of the double-loop design.
ILD achieves significant improvements in final performance, convergence speed, and stability.
arXiv Detail & Related papers (2022-06-10T04:54:32Z) - Accelerated Policy Learning with Parallel Differentiable Simulation [59.665651562534755]
We present a differentiable simulator and a new policy learning algorithm (SHAC)
Our algorithm alleviates problems with local minima through a smooth critic function.
We show substantial improvements in sample efficiency and wall-clock time over state-of-the-art RL and differentiable simulation-based algorithms.
arXiv Detail & Related papers (2022-04-14T17:46:26Z) - DiffSkill: Skill Abstraction from Differentiable Physics for Deformable
Object Manipulations with Tools [96.38972082580294]
DiffSkill is a novel framework that uses a differentiable physics simulator for skill abstraction to solve deformable object manipulation tasks.
In particular, we first obtain short-horizon skills using individual tools from a gradient-based simulator.
We then learn a neural skill abstractor from the demonstration trajectories which takes RGBD images as input.
arXiv Detail & Related papers (2022-03-31T17:59:38Z) - Transformer Inertial Poser: Attention-based Real-time Human Motion
Reconstruction from Sparse IMUs [79.72586714047199]
We propose an attention-based deep learning method to reconstruct full-body motion from six IMU sensors in real-time.
Our method achieves new state-of-the-art results both quantitatively and qualitatively, while being simple to implement and smaller in size.
arXiv Detail & Related papers (2022-03-29T16:24:52Z) - Teaching Robots to Grasp Like Humans: An Interactive Approach [3.3836709236378746]
This work investigates how the intricate task of grasping may be learned from humans based on demonstrations and corrections.
Rather than training a person to provide better demonstrations, non-expert users are provided with the ability to interactively modify the dynamics of their initial demonstration.
arXiv Detail & Related papers (2021-10-09T10:27:50Z) - PlasticineLab: A Soft-Body Manipulation Benchmark with Differentiable
Physics [89.81550748680245]
We introduce a new differentiable physics benchmark called PasticineLab.
In each task, the agent uses manipulators to deform the plasticine into the desired configuration.
We evaluate several existing reinforcement learning (RL) methods and gradient-based methods on this benchmark.
arXiv Detail & Related papers (2021-04-07T17:59:23Z) - UniCon: Universal Neural Controller For Physics-based Character Motion [70.45421551688332]
We propose a physics-based universal neural controller (UniCon) that learns to master thousands of motions with different styles by learning on large-scale motion datasets.
UniCon can support keyboard-driven control, compose motion sequences drawn from a large pool of locomotion and acrobatics skills and teleport a person captured on video to a physics-based virtual avatar.
arXiv Detail & Related papers (2020-11-30T18:51:16Z)
This list is automatically generated from the titles and abstracts of the papers in this site.
This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.