[vc_row row_type=”3″ blox_image=”21431″ align_center=”aligncenter” page_title=”page-title-x” blox_dark=”true” parallax_speed=”8″][vc_column]

---

---

[vc_column_text]

Person/Object tracking in a video stream

[/vc_column_text]

---

---

[/vc_column][/vc_row][vc_row][vc_column width=”1/4″]

---

---

[/vc_column][vc_column width=”1/2″]

---

[vc_column_text]The client wanted to improve its security system. On-premise CCTV video stream as an input in the API.[/vc_column_text]

---

[/vc_column][vc_column width=”1/4″][/vc_column][/vc_row][vc_row][vc_column width=”1/3″][vc_single_image image=”21432″ img_size=”550×350″][/vc_column][vc_column width=”1/3″][vc_single_image image=”21433″ img_size=”550×350″][/vc_column][vc_column width=”1/3″][vc_single_image image=”21434″ img_size=”550×350″][/vc_column][/vc_row][vc_row][vc_column]

---

[vc_column_text]

About this project

[/vc_column_text][vc_row_inner][vc_column_inner width=”1/2″][vc_column_text]Project description

The goal of the project was to create a person, face, and object and a tracking system for a low-resolution video stream. Our solution was integrated into the client’s security system to prevent shoplifting, monitor employees and improve general security.

Fine-tuned triplet based face representation (embedding) on the domain-specific dataset.

[/vc_column_text][/vc_column_inner][vc_column_inner width=”1/2″][vc_column_text]Technologies:

• PyTorch based face detection and vectorization.
• Faiss for fast vector search.
• MongoDB, Postgres for meta-data.

[/vc_column_text][/vc_column_inner][/vc_row_inner]

---

[/vc_column][/vc_row]