LCollision:利用学习的非穿透约束快速生成无碰撞人体姿态
LCollision: Fast Generation of Collision-Free Human Poses using Learned Non-Penetration Constraints
摘要:
我们提出了LCollision,这是一种基于学习的方法,可以合成无碰撞的3D人体姿势。我们的方法的关键是一个新的深层架构,它同时从潜在空间解码新的人类姿势,并预测碰撞的身体部位。这两个组成部分,我们的架构被用来作为目标函数和代理硬约束的约束优化无碰撞的人体姿态生成。我们的方法的一个新颖的方面是使用一个双层自动编码器,分解成局部身体部位之间的碰撞群体的全身碰撞。通过解决约束优化,我们表明,可以解决大量的碰撞文物。此外,在一个大的测试集的$2.5\times 10^6$随机构成SCAPE,我们的架构实现了碰撞预测精度的$94.1\%$与$80\times$加速准确的碰撞检测算法。据我们所知,LCollision是第一种使用神经网络加速碰撞检测和解决渗透问题的方法。
英文摘要:
We present LCollision, a learning-based method that synthesizes collision-free 3D human poses. At the crux of our approach is a novel deep architecture that simultaneously decodes new human poses from the latent space and predicts colliding body parts. These two components of our architecture are used as the objective function and surrogate hard constraints in a constrained optimization for collision-free human pose generation. A novel aspect of our approach is the use of a bilevel autoencoder that decomposes whole-body collisions into groups of collisions between localized body parts. By solving the constrained optimizations, we show that a significant amount of collision artifacts can be resolved. Furthermore, in a large test set of $2.5\times 10^6$ randomized poses from SCAPE, our architecture achieves a collision-prediction accuracy of $94.1\%$ with $80\times$ speedup over exact collision detection algorithms. To the best of our knowledge, LCollision is the first approach that accelerates collision detection and resolves penetrations using a neural network.
机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。