|
INTERNATIONAL JOURNAL OF SCIENTIFIC DEVELOPMENT AND RESEARCH International Peer Reviewed & Refereed Journals, Open Access Journal ISSN Approved Journal No: 2455-2631 | Impact factor: 8.15 | ESTD Year: 2016 open access , Peer-reviewed, and Refereed Journals, Impact factor 8.15 |
Imp Links for Author
Imp Links for Reviewer
Research Area
Subscribe IJSDR
Visitor Counter
Copyright Infringement Claims
Indexing Partner
|
Published Paper Details
|
|
| Paper Title: | Enhancing Referring Expression Segmentation Through Positional Context |
| Authors Name: | Tuba Amreen Darwesh , Anita Harsoor |
| Unique Id: | IJSDR2409030 |
| Published In: | Volume 9 Issue 9, September-2024 |
| Abstract: | Abstract-Referring Expression Segmentation (RES) is a single procedure that involves segmentation construction. mask for an item recognised by a certain linguistic communication. This approach focusses mostly on expressions with a single target, and single expression fits the intended object. Conversely, (GRES) increases the RES task's scope by allowing phrases that indicate any number of targets items. This covers situations with several targets—one target, none at all, and targets. GRES aims to overcome the limitations that RES datasets have, and approaches, hence improving its relevance in a variety of settings.(RES) as well as (GRES) enable improved understanding of visuals by producing human language-based segmentation masks an explanation Using databases such as gRefCOCO and techniques such as ReLA, which bridge the gap between language comprehension as well as computer vision.. The primary objective of RES and GRES is to generate segmentation masks for objects referenced in natural language expressions within images.The aims to enhance image understanding by accurately delineating objects based on linguistic descriptions, thereby improving object recognition and scene interpretation. The implementation of (RES) and (GRES) involves dataset preparation with paired images and language expressions. Neural network models like CNNs or transformer-based architectures are trained on these datasets to learn the relationship between image regions and language descriptions. Trained models generate segmentation masks for referenced objects in new images. Evaluation assesses segmentation accuracy and informs iterative refinement of model architecture. Finally, the optimized model is deployed for applications like as image understanding and human- computer interaction. |
| Keywords: | Index Terms: GRES ,NLP ,Ref Exp Segmentation |
| Cite Article: | "Enhancing Referring Expression Segmentation Through Positional Context", International Journal of Science & Engineering Development Research (www.ijsdr.org), ISSN:2455-2631, Vol.9, Issue 9, page no.276 - 287, September-2024, Available :http://www.ijsdr.org/papers/IJSDR2409030.pdf |
| Downloads: | 00022 |
| Publication Details: | Published Paper ID: IJSDR2409030 Registration ID:212485 Published In: Volume 9 Issue 9, September-2024 DOI (Digital Object Identifier): Page No: 276 - 287 Publisher: IJSDR | www.ijsdr.org ISSN Number: 2455-2631 |
|
Click Here to Download This Article |
|
| Article Preview | |
|
|
|
Major Indexing from www.ijsdr.org
| Google Scholar | ResearcherID Thomson Reuters | Mendeley : reference manager | Academia.edu |
| arXiv.org : cornell university library | Research Gate | CiteSeerX | DOAJ : Directory of Open Access Journals |
| DRJI | Index Copernicus International | Scribd | DocStoc |
Track Paper
Important Links
Conference Proposal
ISSN
 |
 |
DOI (A digital object identifier)
  Providing A digital object identifier by DOI How to GET DOI and Hard Copy Related |
Open Access License Policy
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License
|
 |
 |
 |
Social Media
Indexing Partner |
|||
| Copyright © 2024 - All Rights Reserved - IJSDR | |||
Facebook Twitter Instagram LinkedIn