Chhavi Dhiman
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Chhavi Dhiman has a total of 30 co-authors. Here are the top co-authors by number of shared publications:
- Dhruv Sharma
Department of Electronics and Communication Engineering, Delhi Technological University, Delhi, India
4 shared publications - Dinesh Kant Kumar
School of Engineering, RMIT University, Melbourne, Australia
4 shared publications Aashania Antil
3 shared publications- Aashania Antil
Department of Electronics and Communication Engineering, Delhi Technological University, Delhi, India
3 shared publications - Neha Sharma
Department of Electronics and Communication and Engineering, Delhi Technological University (DTU), Delhi, India
3 shared publications - Dinesh Kumar Vishwakarma
Delhi Technological University
2 shared publications
Latest Publications
- 18 August 2026
- Published by Elsevier in Knowledge-Based Systems
- Vol. 352
Bi-STAM: Bi-directional spatio-temporal adaptive modeling for robust face anti-spoofing
- 18 August 2026
- Published by Elsevier in Knowledge-Based Systems
- Vol. 352
Bi-STAM: Bi-directional spatio-temporal adaptive modeling for robust face anti-spoofing
- 10 June 2026
- Published by Elsevier in Signal Processing: Image Communication
- Vol. 148
A Unified Approach for Multimodal Emotion Recognition Using Counterfactual Learning
- 10 June 2026
- Published by Elsevier in Signal Processing: Image Communication
- Vol. 148
A Unified Approach for Multimodal Emotion Recognition Using Counterfactual Learning
- 21 April 2026
- Published by Springer Nature
- p. 524-532
Graph Based Context-Aware Sentiment Driven Fake News Detection Framework
- 21 April 2026
- Published by Springer Nature
- p. 524-532
Graph Based Context-Aware Sentiment Driven Fake News Detection Framework
- 24 February 2026
- Published by Elsevier in Signal Processing: Image Communication
- Vol. 144
Enhancing memes analysis with multi-task multimodal M2CLIP-SEmoNet architecture
- 24 February 2026
- Published by Elsevier in Signal Processing: Image Communication
- Vol. 144
Enhancing memes analysis with multi-task multimodal M2CLIP-SEmoNet architecture
- 11 October 2025
- Published by Elsevier in Journal of Visual Communication and Image Representation
- Vol. 113
UnMA-CapSumT: Unified and Multi-Head attention-driven caption summarization transformer
- 11 October 2025
- Published by Elsevier in Journal of Visual Communication and Image Representation
- Vol. 113
UnMA-CapSumT: Unified and Multi-Head attention-driven caption summarization transformer
- 5 August 2025
- Published by Elsevier in Neurocomputing
- Vol. 653
Decoding fake news fabrications and trends: A comprehensive survey
- 5 August 2025
- Published by Elsevier in Neurocomputing
- Vol. 653
Decoding fake news fabrications and trends: A comprehensive survey
- 17 June 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Intelligent Transportation Systems
- Vol. 26 (10) , 17110-17120
Cross-Modal Pedestrian Behavior Prediction: A Dual-Task Approach With Progressive Denoising Attention and CVAE
Pedestrian intention and trajectory prediction are crucial for advancing intelligent transportation systems and autonomous vehicles, significantly enhancing urban mobility’s safety and efficiency. Traditional approaches have evolved from capturing pedestrian dynamics through image features and bounding box coordinates to leveraging multiple modalities and attention mechanisms. However, challenges in robust cross-modal feature integration and adaptation to complex scenarios persist. This paper introduces a dual-task approach that simultaneously predicts short-term pedestrian crossing intentions and long-term trajectories by integrating features from pedestrian regions of interest (ROIs), scene attributes, and past trajectories. For crossing intention prediction, Progressive Denoising Attention (PDA) is developed, which iteratively refines cross-modal features to augment inter-class variations. Additionally, a three-phase counterfactual training approach is employed that manipulates pedestrian ROIs and segmentation maps to further enhance model robustness in complex scenarios. For trajectory prediction, a Conditional Variational Autoencoder (CVAE) is implemented, guided by contextual embeddings from the novel Context-Aware Feature Fusion Module (CAFFM) to significantly reduce mean squared error by integrating rich spatiotemporal ROI and context information. Experimental results on benchmark datasets JAAD and PIE demonstrate the superior performance of the proposed approach in understanding and predicting pedestrian intent. The code is available at: https://github.com/neha013/DPITRA
- 17 June 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Intelligent Transportation Systems
- Vol. 26 (10) , 17110-17120
Cross-Modal Pedestrian Behavior Prediction: A Dual-Task Approach With Progressive Denoising Attention and CVAE
Pedestrian intention and trajectory prediction are crucial for advancing intelligent transportation systems and autonomous vehicles, significantly enhancing urban mobility’s safety and efficiency. Traditional approaches have evolved from capturing pedestrian dynamics through image features and bounding box coordinates to leveraging multiple modalities and attention mechanisms. However, challenges in robust cross-modal feature integration and adaptation to complex scenarios persist. This paper introduces a dual-task approach that simultaneously predicts short-term pedestrian crossing intentions and long-term trajectories by integrating features from pedestrian regions of interest (ROIs), scene attributes, and past trajectories. For crossing intention prediction, Progressive Denoising Attention (PDA) is developed, which iteratively refines cross-modal features to augment inter-class variations. Additionally, a three-phase counterfactual training approach is employed that manipulates pedestrian ROIs and segmentation maps to further enhance model robustness in complex scenarios. For trajectory prediction, a Conditional Variational Autoencoder (CVAE) is implemented, guided by contextual embeddings from the novel Context-Aware Feature Fusion Module (CAFFM) to significantly reduce mean squared error by integrating rich spatiotemporal ROI and context information. Experimental results on benchmark datasets JAAD and PIE demonstrate the superior performance of the proposed approach in understanding and predicting pedestrian intent. The code is available at: https://github.com/neha013/DPITRA
- 30 May 2025
- Published by Springer Nature
- p. 157-168
Leveraging Depth Data and Parameter Sharing in Vision Transformers for Improved Face Anti-spoofing
- 30 May 2025
- Published by Springer Nature
- p. 157-168
Leveraging Depth Data and Parameter Sharing in Vision Transformers for Improved Face Anti-spoofing
- 2 December 2024
- Published by Springer Nature
- p. 430-449
Securing Faces: A GAN-Powered Defense Against Spoofing with MSRCR and CBAM
- 2 December 2024
- Published by Springer Nature
- p. 430-449
Securing Faces: A GAN-Powered Defense Against Spoofing with MSRCR and CBAM
- 26 November 2024
- Published by Elsevier in Pattern Recognition
- Vol. 161
Predicting pedestrian intentions with multimodal IntentFormer: A Co-learning approach
- 26 November 2024
- Published by Elsevier in Pattern Recognition
- Vol. 161
Predicting pedestrian intentions with multimodal IntentFormer: A Co-learning approach