Himanshu Sharma
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Himanshu Sharma has a total of 19 co-authors. Here are the top co-authors by number of shared publications:
- Swati Srivastava
GLA University
9 shared publications - Anand Singh Jalal
School of Computer Science & Information Technology, Devi Ahilya Vishwavidyalaya (DAVV), Indore, India
5 shared publications Anand Singh Jalal
GLA University
2 shared publicationsRohit Agarwal
GLA University
2 shared publications- Sonia Singh
Institute of Pharmaceutical Research, GLA University, 17km Stone, NH-2, Mathura-Delhi Road Mathura, Chaumuhan, Uttar Pradesh, 281406, India
2 shared publications Swati Srivastava
Department of Computer Engineering and Applications, GLA University Mathura, Mathura, India
2 shared publications
Latest Publications
- 8 November 2026
- Published by Wiley in Computational Intelligence
- Vol. 42 (6)
TextVQA : A Comprehensive Study of Approaches, Datasets, and Evaluation Metrics
Text‐based visual question answering (Text‐VQA) aims to answer question by combining visual features with embedded textual content. This field has gained substantial attention due to developments in deep learning technologies across vision, text and voice. A typical Text‐VQA pipeline includes Optical Character Recognition (OCR) for extracting textual content from images, which are then employed within a VQA model to reason about the content. In this survey, we conducted a detailed study of the field, systematically analyzing key approaches, datasets, and evaluation metrics that have shaped the expansion of Text‐VQA systems. We provide a comprehensive overview of recent state‐of‐the‐art (SOTA) methods and evaluate their performance on three major datasets: TextVQA, ST‐VQA and OCR‐VQA. By observing each phase of the Text‐VQA methods, we discussed key challenges, acknowledge major contributions, and presented promising directions for future research in this evolving field.
- 8 November 2026
- Published by Wiley in Computational Intelligence
- Vol. 42 (6)
TextVQA : A Comprehensive Study of Approaches, Datasets, and Evaluation Metrics
Text‐based visual question answering (Text‐VQA) aims to answer question by combining visual features with embedded textual content. This field has gained substantial attention due to developments in deep learning technologies across vision, text and voice. A typical Text‐VQA pipeline includes Optical Character Recognition (OCR) for extracting textual content from images, which are then employed within a VQA model to reason about the content. In this survey, we conducted a detailed study of the field, systematically analyzing key approaches, datasets, and evaluation metrics that have shaped the expansion of Text‐VQA systems. We provide a comprehensive overview of recent state‐of‐the‐art (SOTA) methods and evaluate their performance on three major datasets: TextVQA, ST‐VQA and OCR‐VQA. By observing each phase of the Text‐VQA methods, we discussed key challenges, acknowledge major contributions, and presented promising directions for future research in this evolving field.
- 1 March 2025
- Published by SPIE-Intl Soc Optical Eng in Journal of Electronic Imaging
- Vol. 34 (02) , 023012-20
Improving chart question answering through integration of figure captions and visual–textual co-attention
The proposed work for chart question answering (CQA) addresses crucial aspects particularly those associated with questions requiring complex reasoning and visual references to charts and those related to optical character recognition (OCR) noise. Existing methods focused majorly on image features, neglecting the unique characteristics of the visual representation which consequently not able to answer complex reasoning questions. The proposed model employs structural embedding, question embedding, and visual–textual co-attention to address complex reasoning questions, effectively capturing the robust correlation between visual and textual features. The incorporation of the reading component and visual component using bidirectional encoder representations from transformers (BERT) encoding leverages both textual and visual information which overcomes the OCR noise–related issue. Furthermore, the proposed model uses the knowledge acquired during the image captioning task. Moreover, experiments are conducted on five datasets: data visualization question answering (DVQA), FigureQA, PlotQA, LEAFQA++, and ChartQA which makes the assessment robust and generalized. The overall accuracy of our model is increased by 1.17% compared with structure-based transformers with localization for chart question answering (STL-CQA) on Test-Familiar of DVQA, 0.57% compared with PReFIL on FigureQA, 8.16% compared with Plot-QA, 2.33% compared with STL-CQA Test-Familiar Split of LEAFQA++, and 0.79% compared with ChartLlama on ChartQA, which shows the effectiveness and superiority of the proposed model over existing methods on these datasets.
- 1 March 2025
- Published by SPIE-Intl Soc Optical Eng in Journal of Electronic Imaging
- Vol. 34 (02) , 023012-20
Improving chart question answering through integration of figure captions and visual–textual co-attention
The proposed work for chart question answering (CQA) addresses crucial aspects particularly those associated with questions requiring complex reasoning and visual references to charts and those related to optical character recognition (OCR) noise. Existing methods focused majorly on image features, neglecting the unique characteristics of the visual representation which consequently not able to answer complex reasoning questions. The proposed model employs structural embedding, question embedding, and visual–textual co-attention to address complex reasoning questions, effectively capturing the robust correlation between visual and textual features. The incorporation of the reading component and visual component using bidirectional encoder representations from transformers (BERT) encoding leverages both textual and visual information which overcomes the OCR noise–related issue. Furthermore, the proposed model uses the knowledge acquired during the image captioning task. Moreover, experiments are conducted on five datasets: data visualization question answering (DVQA), FigureQA, PlotQA, LEAFQA++, and ChartQA which makes the assessment robust and generalized. The overall accuracy of our model is increased by 1.17% compared with structure-based transformers with localization for chart question answering (STL-CQA) on Test-Familiar of DVQA, 0.57% compared with PReFIL on FigureQA, 8.16% compared with Plot-QA, 2.33% compared with STL-CQA Test-Familiar Split of LEAFQA++, and 0.79% compared with ChartLlama on ChartQA, which shows the effectiveness and superiority of the proposed model over existing methods on these datasets.
- Open Access
- 14 November 2024
- Published by EDP Sciences in E3S Web of Conferences
- Vol. 591, 06003
Hydrogen Production via Electrolyzers: Enhancing Efficiency and Reducing Costs
Hydrogen production through electrolyzers is gaining considerable interest as a sustainable energy solution, particularly in the context of reducing carbon emissions and supporting the global shift towards renewable energy. However, the widespread adoption of hydrogen as an energy carrier is hindered by the relatively low efficiency and high costs of current electrolysis technologies. This paper explores various strategies to enhance the efficiency of electrolyzers, such as advancements in catalyst development, optimization of operational parameters, and innovative system designs. Additionally, it investigates cost-reduction approaches, including improvements in material utilization, integration of renewable energy sources, and economies of scale. A comprehensive analysis is presented, focusing on recent technological advancements, their impact on hydrogen production efficiency, and the potential pathways for making hydrogen a competitive energy solution in the market.
- 14 November 2024
- Published by EDP Sciences in E3S Web of Conferences
- Vol. 591, 06003
Open AccessHydrogen Production via Electrolyzers: Enhancing Efficiency and Reducing Costs
Hydrogen production through electrolyzers is gaining considerable interest as a sustainable energy solution, particularly in the context of reducing carbon emissions and supporting the global shift towards renewable energy. However, the widespread adoption of hydrogen as an energy carrier is hindered by the relatively low efficiency and high costs of current electrolysis technologies. This paper explores various strategies to enhance the efficiency of electrolyzers, such as advancements in catalyst development, optimization of operational parameters, and innovative system designs. Additionally, it investigates cost-reduction approaches, including improvements in material utilization, integration of renewable energy sources, and economies of scale. A comprehensive analysis is presented, focusing on recent technological advancements, their impact on hydrogen production efficiency, and the potential pathways for making hydrogen a competitive energy solution in the market.
- 1 November 2024
- Published by SPIE-Intl Soc Optical Eng in Journal of Electronic Imaging
- Vol. 33 (06) , 063052-18
Deep multimodal relational reasoning and guided attention for chart question answering
The proposed work for chart question answering (CQA) addresses crucial aspects, particularly those associated with questions requiring complex reasoning and visual references to charts and those related to optical character recognition (OCR) noise. Existing methods focused mainly on image features, neglecting the unique characteristics of the visual representation, which consequently not able to answer complex reasoning questions. Introducing structural embedding, question embedding, and a transformer-based relational understanding of the visualizations enable the proposed model to answer complex reasoning questions. The incorporation of the reading component and visual component using BERT encoding leverages both textual and visual information, which overcomes the OCR noise-related issue. The comprehensive ablation study systematically evaluates the different components of our model. Moreover, experiments are conducted on five datasets: DVQA, FigureQA, PlotQA, LEAFQA++, and ChartQA, which makes the assessment robust and generalized. The overall accuracy of our model is increased by 1.15% compared with STL-CQA on the test familiar of DVQA, 0.55% compared with PReFIL on FigureQA, 8.03% compared with Plot-QA, 2.28% compared with STL-CQA test-familiar split of LEAFQA++, and 0.40% compared with VisionTaPas on the test-set of ChartQA, which shows the effectiveness and superiority of the proposed model over existing methods on these datasets.
- 1 November 2024
- Published by SPIE-Intl Soc Optical Eng in Journal of Electronic Imaging
- Vol. 33 (06) , 063052-18
Deep multimodal relational reasoning and guided attention for chart question answering
The proposed work for chart question answering (CQA) addresses crucial aspects, particularly those associated with questions requiring complex reasoning and visual references to charts and those related to optical character recognition (OCR) noise. Existing methods focused mainly on image features, neglecting the unique characteristics of the visual representation, which consequently not able to answer complex reasoning questions. Introducing structural embedding, question embedding, and a transformer-based relational understanding of the visualizations enable the proposed model to answer complex reasoning questions. The incorporation of the reading component and visual component using BERT encoding leverages both textual and visual information, which overcomes the OCR noise-related issue. The comprehensive ablation study systematically evaluates the different components of our model. Moreover, experiments are conducted on five datasets: DVQA, FigureQA, PlotQA, LEAFQA++, and ChartQA, which makes the assessment robust and generalized. The overall accuracy of our model is increased by 1.15% compared with STL-CQA on the test familiar of DVQA, 0.55% compared with PReFIL on FigureQA, 8.03% compared with Plot-QA, 2.28% compared with STL-CQA test-familiar split of LEAFQA++, and 0.40% compared with VisionTaPas on the test-set of ChartQA, which shows the effectiveness and superiority of the proposed model over existing methods on these datasets.
- 29 December 2023
- Published by Springer Nature in Multimedia Tools and Applications
- Vol. 83 (21) , 59577-59595
Integrating multimodal features by a two-way co-attention mechanism for visual question answering
- 29 December 2023
- Published by Springer Nature in Multimedia Tools and Applications
- Vol. 83 (21) , 59577-59595
Integrating multimodal features by a two-way co-attention mechanism for visual question answering
- 12 July 2023
- Published by Springer Nature in Multimedia Tools and Applications
- Vol. 83 (5) , 14085-14109
Face recognition for human identification through integration of complex domain unsupervised and supervised frameworks
- 12 July 2023
- Published by Springer Nature in Multimedia Tools and Applications
- Vol. 83 (5) , 14085-14109
Face recognition for human identification through integration of complex domain unsupervised and supervised frameworks
- 1 May 2023
- Published by SPIE-Intl Soc Optical Eng in Journal of Electronic Imaging
- Vol. 32 (03) , 033023-19
Improving scene text image captioning using transformer-based multilevel attention
Many existing image captioning methods only focus on image objects and their relationships for generating image captions, ignoring the text present in an image. Scene text (ST) contains crucial information to understand an image and facilitating reasoning. The existing methods fail to establish strong correlations between optical character recognition (OCR) tokens, as they have limited OCR representation power. Further, these methods have not efficiently used the positional information of the text. In this work, we have proposed an ST-based image captioning model (Trans-MAtt) based on a multilevel attention mechanism and relation network. We have used relation networks to enhance the connections between ST tokens. We have employed a multi-level attention method, which comprises of spatial, semantic, and appearance attention modules that precisely define the image. To represent context-enriched ST tokens, we use a combination of appearance, location, FastText, and PHOC features. We predict the ST location in the image, which is further integrated with the generated word embeddings for final caption generation. Experiments on the TextCaps dataset demonstrate the effectiveness of the proposed Trans-MAtt model, where it outperforms the current best model by 3.4% on B-4, 2.9% on METEOR, 3.3% on ROUGE-L, 3.1% on CIDEr-D, and 4.1% on SPICE metric scores. Our experiments on the Flickr30k and MSCOCO datasets demonstrated the superiority of our proposed model over existing methods.
- 1 May 2023
- Published by SPIE-Intl Soc Optical Eng in Journal of Electronic Imaging
- Vol. 32 (03) , 033023-19
Improving scene text image captioning using transformer-based multilevel attention
Many existing image captioning methods only focus on image objects and their relationships for generating image captions, ignoring the text present in an image. Scene text (ST) contains crucial information to understand an image and facilitating reasoning. The existing methods fail to establish strong correlations between optical character recognition (OCR) tokens, as they have limited OCR representation power. Further, these methods have not efficiently used the positional information of the text. In this work, we have proposed an ST-based image captioning model (Trans-MAtt) based on a multilevel attention mechanism and relation network. We have used relation networks to enhance the connections between ST tokens. We have employed a multi-level attention method, which comprises of spatial, semantic, and appearance attention modules that precisely define the image. To represent context-enriched ST tokens, we use a combination of appearance, location, FastText, and PHOC features. We predict the ST location in the image, which is further integrated with the generated word embeddings for final caption generation. Experiments on the TextCaps dataset demonstrate the effectiveness of the proposed Trans-MAtt model, where it outperforms the current best model by 3.4% on B-4, 2.9% on METEOR, 3.3% on ROUGE-L, 3.1% on CIDEr-D, and 4.1% on SPICE metric scores. Our experiments on the Flickr30k and MSCOCO datasets demonstrated the superiority of our proposed model over existing methods.
- 22 February 2023
- Published by Taylor & Francis in The Imaging Science Journal
- Vol. 71 (3) , 244-256
Context-aware and co-attention network based image captioning model
- 22 February 2023
- Published by Taylor & Francis in The Imaging Science Journal
- Vol. 71 (3) , 244-256
Context-aware and co-attention network based image captioning model
- 17 December 2022
- Published by Springer Nature in Neural Processing Letters
- Vol. 55 (5) , 5693-5715
A Framework for Image Captioning Based on Relation Network and Multilevel Attention Mechanism
- 17 December 2022
- Published by Springer Nature in Neural Processing Letters
- Vol. 55 (5) , 5693-5715
A Framework for Image Captioning Based on Relation Network and Multilevel Attention Mechanism
- 16 September 2022
- Published by Springer Nature in Multimedia Tools and Applications
- Vol. 82 (7) , 10981-11003
Multilevel attention and relation network based image captioning model
- 16 September 2022
- Published by Springer Nature in Multimedia Tools and Applications
- Vol. 82 (7) , 10981-11003
Multilevel attention and relation network based image captioning model