Publications & Papers
Peer-reviewed work across computer vision, time-aware retrieval, and applied machine learning, published at major conferences and in academic journals.
Conference and Journal Publications
TIGeR: A Unified Framework for Time, Images and Geo-location Retrieval
IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) 2026 — David G. Shatwell, Sirnam Swetha, Mubarak Shah
Applications like digital forensics, urban monitoring, and environmental analysis frequently require reasoning over appearance, place, and time simultaneously. This paper frames that problem as Geo-Time Aware Image Retrieval and builds a benchmark of 4.5M image–location–time triplets for training and 85k triplets for evaluation. TIGeR is a multimodal transformer that maps images, geolocation, and time into one shared geo-temporal embedding space.
Project Page ArxivVRR-QA: Visual Relational Reasoning in Videos Beyond Explicit Cues
IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) 2026 — Sirnam Swetha, Rohit Gupta, Parth Parag Kulkarni, David G. Shatwell, Jeffrey A. Chan Santiago, Nyle Siddiqui, Joseph Fioresi, Mubarak Shah
Most existing VideoQA benchmarks focus on content that is plainly visible within individual frames. To address the gap in implicit reasoning across cinematic cuts and implied relationships, this work presents VRR-QA, an expert-annotated benchmark drawn from creative storytelling videos that reveals clear performance drops in current models.
GT-Loc: Unifying When and Where in Images Through a Joint Embedding Space
International Conference on Computer Vision, ICCV 2025 — Oral Presentation (top 0.6%) — David G. Shatwell, Ishan Rajendrakumar Dave, Sirnam Swetha, Mubarak Shah
GT-Loc introduces a retrieval-based framework that estimates hour, date, and GPS coordinates together using cyclical toroidal representations and unified embedding spaces, outperforming earlier time prediction baselines even when ground-truth locations are provided.
Project Page ArxivReal Time Ore Sorting Using Color and Texture Analysis
International Journal of Mining Science and Technology, 2023 — David Shatwell, Victor Murray, Augusto Barton
This sensor-based ore sorting method combines wavelet texture features, color statistics, and neural classifiers to separate high-grade ore from waste rock while maintaining processing latency below 44 ms.
PaperFrequently Asked Questions
© 2026 David Shatwell
Each entry above links to the project page, arXiv preprint, code, or dataset where one is available, so you can dig into the methods and results directly. The collection reflects a consistent focus on reasoning over time, place, and appearance, from geo-temporal retrieval to video question answering and industrial imaging.
