A comprehensive hands-on course developed with Arm University covering optimization of generative AI workloads across Arm architectures. Features three practical labs spanning mobile devices, cloud servers, and edge-to-cloud deployment comparisons using Raspberry Pi 5 and AWS Graviton.
Test your own place-recognition skills against state-of-the-art open- and closed-source VLMs in this interactive GeoGuessr-style game.
A compact and efficient approach for Visual Place Recognition using extreme low-bit quantization and progressive distillation.
VSLAMs a streamlined Python implementation of Stereo Visual SLAM. It leverages libraries such as numpy, opencv, and scipy for feature detection, tracking, matching, motion estimation, and optimization—all designed with the KITTI dataset in mind.
An adaptive ternary-quantized Vision Transformer that lets robots dial accuracy versus compute on the fly—shrinking model size 5 × and cutting inference operations by up to 40 % while maintaining state-of-the-art Recall @ 1.
Exploring the growing privacy risks posed by Vision-Language Models that can pinpoint locations from images, and our research into responsible geolocation AI.
Published in IEEE Robotics and Automation Letters, 2024
This paper examines the impact of compact architectural designs and quantization techniques on visual place recognition, balancing recall performance with memory and latency constraints for edge deployment.
Recommended citation: Grainge, O., Milford, M., Bodala, I., Ramchurn, S. D., & Ehsan, S. (2024). "Design Space Exploration of Low-Bit Quantized Neural Networks for Visual Place Recognition." IEEE Robotics and Automation Letters. 9(6), 5070-5077. doi:10.1109/LRA.2024.3386459
Download Paper | Download Slides
Published in IEEE Robotics and Automation Letters, 2024
This paper introduces a structured pruning method for VPR that strategically removes redundancies in the embedding space, significantly reducing memory usage and latency with minimal impact on accuracy.
Recommended citation: Grainge, O., Milford, M., Bodala, I., Ramchurn, S. D., & Ehsan, S. (2025). "Structured Pruning for Efficient Visual Place Recognition." IEEE Robotics and Automation Letters, 10(2), 2024-2031. doi:10.1109/LRA.2024.3523222
Download Paper | Download Slides
Published in arXiv preprint, 2025
This paper introduces TeTRA, a ternary transformer approach that progressively quantizes Vision Transformers to achieve significant reductions in memory consumption and inference latency, while preserving or even enhancing visual place recognition performance on resource-constrained platforms.
Recommended citation: Grainge, O., Milford, M., Bodala, I., Ramchurn, S. D., & Ehsan, S. (2025). "TeTRA-VPR: A Ternary Transformer Approach for Compact Visual Place Recognition." arXiv preprint, arXiv:2503.02511. doi:10.48550/arXiv.2503.02511
Download Paper | Download Slides
Published in arXiv preprint, 2025
TAT-VPR fuses ternary weight quantization with a learned activation-sparsity gate, giving visual SLAM systems a 5 × smaller model and up to 40 % fewer operations while retaining state-of-the-art Recall@1.
Recommended citation: Grainge, O., Milford, M., Bodala, I., Ramchurn, S. D., & Ehsan, S. (2025). "TAT-VPR: Ternary Adaptive Transformer for Dynamic and Efficient Visual Place Recognition." arXiv preprint, arXiv:2505.16447.
Download Paper | Download Slides
Published:
This is a description of your talk, which is a markdown file that can be all markdown-ified like any other post. Yay markdown!
Published:
This is a description of your conference proceedings talk, note the different field in type. You can put anything in this field.
Undergraduate course, University 1, Department, 2014
This is a description of a teaching experience. You can use markdown like any other post.
Workshop, University 1, Department, 2015
This is a description of a teaching experience. You can use markdown like any other post.