Invariance, Geometry and Deep Neural Networks with Pavan Turaga - #386
The TWIML AI Podcast (formerly This Week in Machine Learning & Artificial Intelligence)
00:00
Enhancing Activity Classification with Time Constraints
This chapter delves into benchmark datasets for motion capture and activity classification, including notable examples like Microsoft's RGBD and NTU. It also highlights advancements in activity detection using time series models, particularly LSTMs and CNNs, and discusses the positive impact of a time-warping constraint on model performance.
Transcript
Play full episode
