Project

General

Profile

Proposed systems » History » Version 2

Version 1 (Petri Partanen, 13/05/2020 20:16) → Version 2/3 (Petri Partanen, 13/05/2020 20:24)

h1. Proposed systems

h2. First proposal
!first_proposal!
The system is based on the following papers:
* Automatic Health Problem Detection from Gait Videos Using Deep Neural Networks
* Stacked Hourglass Networks for Human Pose Estimation
* A simple yet effective baseline for 3d human pose estimation
* Multivariate LSTM-FCNs for Time Series Classification

Pros:
* Based on latest state-of-the-art

Cons:
* Uses neural networks
* Time-consuming to implement and test