Completed
PLAYER AND BALL CAN BE DETECTED PER FRAME
Class Central Classrooms beta
YouTube videos curated by Class Central.
Classroom Contents
Video Analytics for Football Games
Automatically move to the next video in the Classroom when playback concludes
- 1 Intro
- 2 PLAYER AND BALL CAN BE DETECTED PER FRAME
- 3 EVENT DETECTION REQUIRES SEQUENCE OF FRAMES
- 4 PROJECT CONTEXT
- 5 THE PROBLEM LANDSCAPE
- 6 TWO SOLUTION PARTS
- 7 THE DATA FACTS
- 8 LEVERAGE THE MODEL TO SPEED UP THE LABELING
- 9 THE MODELS
- 10 PLAYER DETECTION: FIELD TRANSFORM
- 11 A WALK IN FEATURE SPACE
- 12 SUBTRACT BACKGROUND TO REMOVE THE NOISE
- 13 COORDINATES AS UNLOCKED DOWNSTREAM FEATURE
- 14 START OF GAME MODEL BEATS THE OTHER GOAL MODELS (FOR NOW)
- 15 SOLUTION ARCHITECTURE
- 16 ABOUT APACHE BEAM
- 17 THE SOLUTION LANDSCAPE
- 18 FROM HLS TO JPEG
- 19 FULLY LEVERAGE MANAGED SERVICES
- 20 LEVERAGE THE BEAM MODEL FOR PROCESSING
- 21 WHERE THE DATA CRUNCHING HAPPENS
- 22 PIPELINE DEEP DIVE
- 23 LEVERAGE THE INTERNAL LOAD BALANCER OF GKE TO GET PREDICTIONS
- 24 DEWARPING THE BOUNDING BOXES TO GET COORDINATES
- 25 TEAM DETECTION WITHOUT BACKGROUND SUBTRACTION
- 26 DUMPING THE PREDICTIONS TO BIGTABLE
- 27 LEVERAGE THE BEAM MODEL TO WINDOW THE DATA
- 28 RESPECT THE BEAM MODEL TO GET DESIRED PARALLELIZATION
- 29 TEST IN STREAM MODE
