| I&T Solution |
Video Analytics Solution on Public Columbaria and Public Markets
(REF: S-0783) |
| Trial Project |
|
| Solution Feature |
- Engagement categorized patrons counting - Patrons are detected and counted in camera feed. Patrons will be further categorized into engagement types following configurable rules, including designated area and minimum stay time, according to different facility and environmental needs.
- In/out patrons counting - In/out line can be configured for each camera. A person passing through an in/out line in opposite directions will be counted as entering/leaving a facility.
- Demographics detection - Images of the detected patron from camera feed will be further analyzed to extract demographic information, including age group and gender. Such information will be integrated into statistical analysis.
- Analysis will be conducted on the collected engagement categorized and demographically featured data. Analysis includes time-based statistical analysis and flow pattern analysis and visualization.
- Integration - system integration to communicate real-time and analysis results with other system components in the local network.
|
| Trial Application and Expected Outcome |
- Select a few trial deployment sites (e.g. one site per each selected facility type including Public Columbaria, Garden of Remembrance (GoR) and Public Markets). Sample videos for each site will be required from FEHD for a better understanding of the data, environments and potential difficulty.
- Configuration and AI training: Communicate with FEHD and analyze the requirements for each of the sites and configure the system accordingly. Use a portion of the videos for AI model training and finetuning.
- Testing and deliverables: Test the AI model with the remaining videos. Deliverables include visualization of real-time counting and demographics detection results, time-based statistical analysis CSV results, and flow pattern analysis visualization. An accuracy report will be provided.
- Outcome: Requirements consolidation, accuracy estimation, proof of concept.
|
| Additional Solution Information |
datax_video analytics_sep1_rs.pdf
|
| Info on I&T Solution Provider |
| Solution Provider | : | Datax Limited | | Address | : | Lion Rock 72, 1/F, Innocentre, 72 Tat Chee Ave, Kowloon Tong | | Contact Person | : | Aric Fung |
| Position | : | Account Director | | Tel | : | 91438473 | | Email | : |
aric.fung@datax.io | | Webpage | : | https://datax.io |
|
