.png){kind=small}
OVERVIEW
Estimate the level of coordination between 2 dancers dancing the same choreography
-
According to numerous studies and reports, procurement is the single most time-consuming and expensive stage in a supply chain
-
Some even say that optimizing this stage through digitalization could help companies increase their profits by USD 60000 to 100000!
-
Therefore, my team and I decided to conduct user research with important stakeholders from the procurement stage to understand if digitalization and in particular, chatbots would have a significant impact on this stage
PROJECT DETAILS
Organization
CISCO – RVCE Center of Excellence in Internet of Things
Duration
1 month
Role
Developer
PROBLEM STATEMENT
“ To understand the relevance, need for and to develop chatbots for the ease of procurement stage in the supply chain ”
DESK RESEARCH
Utilized pre-trained model MediaPipe
-
MediaPipe is a popular pre-trained model for human pose estimation as this topology consists of 33 human body keypoints, which is a superset of COCO, BlazeFace and BlazePalm topologies.
-
This allows us to determine body semantics from pose prediction alone that is consistent with face and hand models.
DEVELOPMENT OF TRACKER
Check your accuracy of a choreography or the coordination of dancers in a group
-
This application was programmed using Python and 2 files were used
-
dance_tracker.py
-
Loads the pose estimation module from MediaPipe as objects
-
Preprocesses frames in the input video in order to detect pose landmarks accurately
-
Calculates and returns a set of measurements used to determine similarity between input videos
-
Displays videos side by side for user’s reference
-
-
comparison_main.py
-
Reads input data
-
Calls and utilizes services from dance_tracker.py for detection of pose points
-
Plots results obtained for effective analysis
-
Check out the code here:
✦ More Projects ✦
Thanks for visiting the COOP!
Want to strike up an egg-citing conversation?
Reach out to me via the following links:
