Vijay Kumar B G

I am a Senior Researcher at NEC Laboratories, America. Before joining NEC, I was a Research Scientist at PARC and prior to that I was a Research Fellow at the Australian Centre for Robotic Vision working with Prof. Ian Reid and Dr. Gustavo Carneiro. Before joining ACRV, I was a Researcher at the Advanced Research Group, Samsung Research. I completed my Ph.D. in Computer science from Queen Mary University of London under Prof. Ioannis Patras. I received my MS in Electrical Engineering from IIT Madras.

Email / Scholar / linkedin

Research

I am interested in machine learning and computer vision, with current focus on LLM Agents, multimodal large language models, vision-language understanding, and self-supervised representation learning.

Selected Publications
	Self-Training Large Language Models for Improved Visual Program Synthesis With Visual Reinforcement Zaid Khan, Vijay Kumar B G, Samuel Schulter, Yun Fu, Manmohan Chandraker. IEEE International Conference on Computer Vision and Pattern Recognition (CVPR), 2024 Project page / Paper / Arxiv
	Generating Enhanced Negatives for Training Language-Based Object Detectors Shiyu Zhao, Long Zhao, Vijay Kumar B G, Yumin Suh, Dimitris Metaxas, Manmohan Chandraker, Samuel Schulter. IEEE International Conference on Computer Vision and Pattern Recognition (CVPR), 2024 Code / Arxiv
	Taming Self-Training for Open-Vocabulary Object Detection Shiyu Zhao, Samuel Schulter, Long Zhao, Zhixing Zhang, Vijay Kumar B G, Yumin Suh, Manmohan Chandraker, Dimitris Metaxas. IEEE International Conference on Computer Vision and Pattern Recognition (CVPR), 2024 Code / Arxiv
	LLM-Assist: Enhancing Closed-Loop Planning with Language-Based Reasoning S P Sharan, Francesco Pittaluga, Vijay Kumar B G, Manmohan Chandraker. Arxiv, 2023 Project page / Video / Arxiv
	Exploring Question Decomposition for Zero-shot VQA Zaid Khan, Vijay Kumar B G, Samuel Schulter, Yun Fu, Manmohan Chandraker. Neural Information Processing Systems (NeurIPS), 2023 project page / Arxiv
	DP-Mix: Mixup-based Data Augmentation for Differentially Private Learning Wenxuan Bao, Francesco Pittaluga, Vijay Kumar B G, Vincent Bindschaedler. Neural Information Processing Systems (NeurIPS), 2023 Code / Paper
	OmniLabel: A Challenging Benchmark for Language-Based Object Detection Samuel Schulter, Vijay Kumar B G, Yumin Suh, Konstantinos M. Dafnis, Zhixing Zhang, Shiyu Zhao, Dimitris Metaxas. IEEE Conference on Computer Vision (ICCV), 2023 (Oral: acceptance rate < 3%) Project page / Video / Arxiv
	Q: How to Specialize Large Vision-Language Models to Data-Scarce VQA Tasks? A: Self-Train on Unlabeled Images! Zaid Khan, Vijay Kumar B G, Samuel Schulter, Xiang Yu, Yun Fu, Manmohan Chandraker. IEEE International Conference on Computer Vision and Pattern Recognition (CVPR), 2023 Code / Arxiv
	Single-Stream Multi-Level Alignment for Vision Language Pretraining Zaid Khan, Vijay Kumar B G, Samuel Schulter, Xiang Yu, Yun Fu, Manmohan Chandraker. European Conference on Computer Vision (ECCV), 2022 Code / Arxiv
	Exploiting Unlabeled Data with Vision and Language Models for Object Detection Shiyu Zhao, Zhixing Zhang, Samuel Schulter, Long Zhao, Vijay Kumar B G, Anastasis Stathopoulos, Manmohan Chandraker, Dimitris Metaxas. European Conference on Computer Vision (ECCV), 2022 Code / Arxiv
	STRIVE: Scene Text Replacement In Videos Vijay Kumar B G, Jeyasri Subramanian, Varnith Chordia, Eugene Bart, Shaobo Fang, Kelly Guan, Raja Bala. IEEE Conference on Computer Vision (ICCV), 2021 Data / Project page / Arxiv
	A Theoretically Sound Upper Bound on the Triplet Loss for Improving the Efficiency of Deep Distance Metric Learning Thanh-Toan Do, Toan Tran, Ian Reid, Vijay Kumar B G, Tuan Hoang, Gustavo Carneiro. IEEE International Conference on Computer Vision and Pattern Recognition (CVPR), 2019 Arxiv
	Multi-modal Cycle-consistent Generalized Zero-Shot Learning Rafael Felix, Vijay Kumar B G, Ian Reid, Gustavo Carneiro. European Conference on Computer Vision (ECCV), 2018 Code / Arxiv
	Bayesian Semantic Instance Segmentation in Open Set World Trung Pham, Vijay Kumar B G, Thanh-Toan Do, Gustavo Carneiro, and Ian Reid. European Conference on Computer Vision (ECCV), 2018 Code / Arxiv
	Cartman: The low-cost Cartesian Manipulator that won the Amazon Robotics Challenge D Morrison et. al. IEEE International Conference on Robotics and Automation (ICRA), 2018 Media / Arxiv
	Smart Mining for Deep Metric Learning Vijay Kumar B G, Ben Harwood, Gustavo Carneiro, Ian Reid, and Tom Drummond. IEEE Conference on Computer Vision (ICCV), 2017 Arxiv
	DeepSetNet: Predicting Sets with Deep Neural Networks Seyed Hamid Rezatofighi, Vijay Kumar B G, Anton Milan, Ehsan Abbasnejad, Antony Dick, Ian Reid. IEEE Conference on Computer Vision (ICCV), 2017 (Spotlight: acceptance rate < 5%) Arxiv
	Unsupervised CNN for Single View Depth Estimation: Geometry to the Rescue Ravi Garg, Vijay Kumar B G, Gustavo Carneiro, and Ian Reid. European Conference on Computer Vision (ECCV), 2016 Code / Arxiv
	Learning Local Image Descriptors with Deep Siamese and Triplet Convolutional Networks by Minimizing Global Loss Functions Vijay Kumar B G, Gustavo Carneiro, and Ian Reid. IEEE International Conference on Computer Vision and Pattern Recognition (CVPR), 2016 (Spotlight: acceptance rate < 10%) Code / Arxiv

Template Credits Jon Barron

Research

Selected Publications