An image marking system and a method thereof
The image masking system addresses the challenge of user intent interpretation in image search by segmenting and masking irrelevant areas, improving accuracy and reducing computational overhead in tasks like autonomous driving and surveillance.
DE102025149222A1Pending Publication Date: 2026-05-28BOSCH GLOBAL SOFTWARE TECH PTE LTD +1
View PDF 0 Cites 0 Cited by
Patent Information
- Application Number
- DE102025149222
- Authority / Receiving Office
- DE · DE
- Patent Type
- Applications
- Current Assignee / Owner
- Priority Date
- 2024-11-28
- Filing Date
- 2025-11-26
- Publication Date
- 2026-05-28
AI Technical Summary
Technical Problem
Existing image search systems fail to interpret user intent and focus on specific areas of interest, leading to irrelevant results and increased computational inefficiency due to processing of all image objects equally.
Method used
An image masking system that segments and masks irrelevant areas using a convolutional neural network and/or transformer architecture, generating a binary mask to isolate areas of interest and extract embeddings.
Benefits of technology
Improves accuracy and reduces computational overhead by focusing on user-specified areas, enhancing performance in tasks like autonomous driving and surveillance by isolating relevant features and reducing noise.
✦ Generated by Eureka AI based on patent content.
Smart Images
Figure 00000000_0000_ABST
Abstract
An image masking system (100) is disclosed. The system (100) includes a segmentation module (130) configured to extract one or more areas of interest by processing an input query image using a segmentation mesh. The segmentation module (130) is also configured to select one or more segments from one or more extracted areas of interest based on a user indicator. The system (100) includes a mask generation module (140) configured to generate a binary mask for selected segments on the input query image. The system (100) includes an image transformer module (150) configured to generate a masked image by performing an arithmetic operation on the binary mask and the input query image.The image transformer module (150) is also configured to extract embeddings of the masked image using a convolutional neural network or transformer architecture.
Need to check novelty before this filing date? Find Prior Art