Trishna Paul
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Trishna Paul has a total of 11 co-authors. Here are the top co-authors by number of shared publications:
- Mukesh Kumar Rohil
Department of Computer Science and Information Systems, Birla Institute of Technology and Science Pilani, Pilani Campus, Pilani, India
3 shared publications - Arindam Kolay
Department of Pharmaceutics, NIMS Institute of Pharmacy, NIMS University, Rajasthan, Jaipur, India
2 shared publications A. K. Singh
Department of Mathematics, Banaras Hindu University, Varanasi, India
1 shared publication- Ajay Kumar Mishra
Department of Chemistry, University of Western Cape, Bellville, Cape Town 7535, South Africa
1 shared publication Alka Beniwal
Department of Computer Science and Information Systems, Birla Institute of Technology and Science Pilani, Pilani, India
1 shared publicationArindam Kolay
NIMS University Rajasthan,NIMS Institute of Pharmacy,Jaipur,India
1 shared publication
Latest Publications
- 3 February 2026
- Published by Springer Nature
- p. 479-494
Multi-agent Decision Support Framework for Bed Allocation and Patient Flow Optimization in Hospitals
- 3 February 2026
- Published by Springer Nature
- p. 479-494
Multi-agent Decision Support Framework for Bed Allocation and Patient Flow Optimization in Hospitals
- 1 October 2025
- Published by Springer Nature
- p. 297-309
Advancements and Applications of 3D Bioprinted Scaffold in Diabetic Wound Healing
- 1 October 2025
- Published by Springer Nature
- p. 297-309
Advancements and Applications of 3D Bioprinted Scaffold in Diabetic Wound Healing
- Open Access
- 26 September 2025
- Published by Springer Nature in Scientific Reports
- Vol. 15 (1) , 1-13
Improved 3D reconstruction pipeline for enhancing the quality of 3D model generation from multi-view images
In this research, we slightly modify the conventional pipeline such that it (1) sets a minimum triangulation angle of 3°, (2) reduces (and possibly minimizes) overall re-projection error by simultaneously optimizing all camera poses and 3D points in the bundle adjustment step, and (3) uses a tiling buffer size of 1024 × 1024, to generate, from a set of 2D images, a detailed 3D models of complex objects. We systematically showcase the versatility of this improved photogrammetric pipeline by applying it to three datasets - a proprietary dataset (money plant) and two publicly available datasets (a statue and an old computer). To the best of our knowledge, while previous research has employed photogrammetry techniques for 3D model generation, none has systematically defined the 3d reconstruction pipeline, with a focus on underlying mathematical concepts, Additionally, there has been a lack of evaluation of how the quality of input 2D images impacts quality of the resulting 3D models. Our study evaluates the entropy and image quality (using BRISQUE scores) of the 3D model generated in relation to the image quality of the set of input 2D images, providing insights into the reliability of the proposed photogrammetric approach. Despite utilizing a low-quality image dataset as input, we demonstrated that we achieved a high-quality 3D model. Furthermore, we observed less differences in image quality when comparing the 3D model with the original 2D images. Overall, our research contributes to both theoretical understanding and practical application of photogrammetry for 3D reconstruction as experimented on three distinct image datasets. As a future research scope, we provide suggestions for possible parallelization of some of the steps of the 3D reconstruction pipeline.
- 26 September 2025
- Published by Springer Nature in Scientific Reports
- Vol. 15 (1) , 1-13
Open AccessImproved 3D reconstruction pipeline for enhancing the quality of 3D model generation from multi-view images
In this research, we slightly modify the conventional pipeline such that it (1) sets a minimum triangulation angle of 3°, (2) reduces (and possibly minimizes) overall re-projection error by simultaneously optimizing all camera poses and 3D points in the bundle adjustment step, and (3) uses a tiling buffer size of 1024 × 1024, to generate, from a set of 2D images, a detailed 3D models of complex objects. We systematically showcase the versatility of this improved photogrammetric pipeline by applying it to three datasets - a proprietary dataset (money plant) and two publicly available datasets (a statue and an old computer). To the best of our knowledge, while previous research has employed photogrammetry techniques for 3D model generation, none has systematically defined the 3d reconstruction pipeline, with a focus on underlying mathematical concepts, Additionally, there has been a lack of evaluation of how the quality of input 2D images impacts quality of the resulting 3D models. Our study evaluates the entropy and image quality (using BRISQUE scores) of the 3D model generated in relation to the image quality of the set of input 2D images, providing insights into the reliability of the proposed photogrammetric approach. Despite utilizing a low-quality image dataset as input, we demonstrated that we achieved a high-quality 3D model. Furthermore, we observed less differences in image quality when comparing the 3D model with the original 2D images. Overall, our research contributes to both theoretical understanding and practical application of photogrammetry for 3D reconstruction as experimented on three distinct image datasets. As a future research scope, we provide suggestions for possible parallelization of some of the steps of the 3D reconstruction pipeline.
- 29 July 2025
- Published by Springer Nature
- p. 647-661
Virtual Labs and Augmented Reality Applications in Work-Integrated Learning: A Comparison, Challenges, and Insights
- 29 July 2025
- Published by Springer Nature
- p. 647-661
Virtual Labs and Augmented Reality Applications in Work-Integrated Learning: A Comparison, Challenges, and Insights
- 14 August 2024
- Published by Springer Nature in Education and Information Technologies
- Vol. 30 (3) , 3279-3308
An architecture to intertwine augmented reality and intelligent tutoring systems: towards realizing technology-enabled enhanced learning
- 14 August 2024
- Published by Springer Nature in Education and Information Technologies
- Vol. 30 (3) , 3279-3308
An architecture to intertwine augmented reality and intelligent tutoring systems: towards realizing technology-enabled enhanced learning
- 21 February 2024
- Published by Springer Nature
- p. 453-463
Advanced Intelligent Tutoring Systems: Featuring the Learning Management Systems of the Future
- 21 February 2024
- Published by Springer Nature
- p. 453-463
Advanced Intelligent Tutoring Systems: Featuring the Learning Management Systems of the Future
- 7 December 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 631-636
Exploring Breast Cancer Detection Techniques Using Artificial Intelligence
Cancer is one of the most hazardous illnesses for people, but no long-term treatment is currently available. Female mortality rate is high due to breast cancer, it can be reduced by early detection and early treatment. According to the National Breast Cancer Foundation an estimated 297,790 women and 2,800 men in the United States will be diagnosed with invasive breast cancer in 2023. Modern clinical detection tools like Ultra-Sound, Mammography, and Magnetic Resonance Imaging are essential for breast cancer detection. The efficient use of artificial intelligence (AI) and machine learning (ML) in detecting and treating several deadly diseases has increased patient survival rates by facilitating early diagnosis and treatment. In this paper, we discussed about various AI-based breast cancer screening methods and how they operate. The researcher will be helped by this review to gain insight into the challenges and advancement of breast cancer detection technology.
- 7 December 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 631-636
Exploring Breast Cancer Detection Techniques Using Artificial Intelligence
Cancer is one of the most hazardous illnesses for people, but no long-term treatment is currently available. Female mortality rate is high due to breast cancer, it can be reduced by early detection and early treatment. According to the National Breast Cancer Foundation an estimated 297,790 women and 2,800 men in the United States will be diagnosed with invasive breast cancer in 2023. Modern clinical detection tools like Ultra-Sound, Mammography, and Magnetic Resonance Imaging are essential for breast cancer detection. The efficient use of artificial intelligence (AI) and machine learning (ML) in detecting and treating several deadly diseases has increased patient survival rates by facilitating early diagnosis and treatment. In this paper, we discussed about various AI-based breast cancer screening methods and how they operate. The researcher will be helped by this review to gain insight into the challenges and advancement of breast cancer detection technology.
- 1 December 2002
- Published by Springer Nature in Forschung im Ingenieurwesen
- Vol. 67 (5) , 198-205
Mixed convection flow in a porous medium bounded by two vertical walls
- 1 December 2002
- Published by Springer Nature in Forschung im Ingenieurwesen
- Vol. 67 (5) , 198-205
Mixed convection flow in a porous medium bounded by two vertical walls