features_extraction.md

Tutorial 2: Features Extraction

This tutorial will help you, step-by-step, how to extract features (Angles and Trajectories) from 2D Poses.

Before proceeding make sure that you have already extract or download 2D poses, see 2D Poses Extraction for more information.

Experiments were performed in two public dataset KTH and Weizmann.

In this example, we will extract features from 2D poses by running the following scripts.

Computing Angles Features

KTH dataset example:

python tools/FeaturesExtraction/compute_angles_from_body_parts.py \
--poses_base_dir=/home/murilo/dataset/KTH \
--input_dir=2DPoses_Person \
--output_dir=Angles_from_2DPoses

Weizmann dataset example:

python tools/FeaturesExtraction/compute_angles_from_body_parts.py \
--poses_base_dir=/home/murilo/dataset/Weizmann \
--input_dir=2DPoses \
--output_dir=Angles_from_2DPoses

Computing Trajectories Features

KTH dataset example:

python tools/FeaturesExtraction/compute_trajectory_from_body_parts.py \
--poses_base_dir=/home/murilo/dataset/KTH \
--input_dir=2DPoses_Person \
--output_dir=Trajectories_from_2DPoses \
--number_frames=20 --stride=10

Weizmann dataset example:

python tools/FeaturesExtraction/compute_trajectory_from_body_parts.py \
--poses_base_dir=/home/murilo/dataset/Weizmann \
--input_dir=2DPoses \
--output_dir=Trajectories_from_2DPoses \
--number_frames=20 --stride=1

Next

As next step follow the link: Human Action Recognition