[vc_row row_type=”3″ blox_image=”21426″ align_center=”aligncenter” page_title=”page-title-x” blox_dark=”true” parallax_speed=”8″][vc_column]

---

---

[vc_column_text]

Content-Based Image Retrieval

[/vc_column_text]

---

---

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

---

---

[/vc_column][vc_column width=”1/2″]

---

[vc_column_text]An alcoholic beverage industry company wanted to implement new innovative channels of selling its products all over the world. Their task was to search for products using bottle labels from a mobile app.[/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=”21427″ img_size=”550×350″][/vc_column][vc_column width=”1/3″][vc_single_image image=”21428″ img_size=”550×350″][/vc_column][vc_column width=”1/3″][vc_single_image image=”21429″ 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

Label extraction from raw images and content-based image retrieval with deep convolutional neural network + keypoint detection and comparison with SIFT. Self-supervised approach for features adjustment.

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

• PyTorch based CNN models for label segmentation and image representation.
• Faiss for fast vector metric search.
• MongoDB for meta-information storing.

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

---

[/vc_column][/vc_row]