We tackle fine-grained indoor localization from natural language: given a free-form description of one’s surroundings, estimate the observer’s 2D position and heading within a known 3D environment. Language queries are lightweight, privacy-preserving, and need no camera - yet prior work stops at coarse scene retrieval and cannot resolve an intra-scene pose. We close this gap with LangLoc, a three-stage pipeline that (i) retrieves the correct scene via a dual-branch GATv2 encoder with CLIP semantic features, surpassing the previous best by 8 percentage points in Top-1 recall; (ii) estimates position and heading by scoring a dense floor grid through ray-cast object visibility, reaching a median error of 0.95 m; and (iii) resolves residual ambiguity through a Bayesian dialog module that asks targeted yes/no questions and updates a pose posterior until the location is pinpointed. To support this task we contribute a benchmark of 13,000+ pose-indexed natural-language descriptions over 1,300+ indoor 3D scans.
@inproceedings{panwar2026langloc,title={LangLoc: ``Tell Me What You See''},author={Panwar, Shaurya Kishore and Nobari, Roham Zendehdel and Lau, Shirley Feng Yi and Shaik, Abu Bakr Rahman and G{\"u}nther, Manuel and Pollefeys, Marc and Barath, Daniel},booktitle={European Conference on Computer Vision (ECCV)},year={2026},doi={10.48550/arXiv.2607.05077},}
2025
ICICCT
Elevating Aerial Vision-Language Navigation with Enhanced Feature Extraction Backbone Using GELAN and PGI
Shaurya Kishore Panwar, Ajeet Kumar, and Anil Singh Parihar
In Intelligent Computing and Communication Techniques: Proceedings of the International Conference on Intelligent Computing and Communication Techniques (ICICCT 2024), New Delhi, India, 28–29 June, 2024 (Volume 2) , Jun 2025
Aerial Vision and Dialog Navigation (AVDN) is an emerging field in multimodal learning and vision-language understanding. We introduce a new feature-extraction backbone using Generalized Efficient Layer Aggregation Network (GELAN) and Programmable Gradient Information (PGI) to capture more comprehensive visual context. The improved backbone uses a dual-path strategy that optimizes gradient flow and feature reuse, improving learning efficiency and inference speed without excessive computational cost. Our model outperforms SOTA methods on the Aerial Vision Navigation from Dialog History (ANDH) task across multiple metrics.
@incollection{panwar2025aerial,title={Elevating Aerial Vision-Language Navigation with Enhanced Feature Extraction Backbone Using {GELAN} and {PGI}},author={Panwar, Shaurya Kishore and Kumar, Ajeet and Parihar, Anil Singh},booktitle={Intelligent Computing and Communication Techniques: Proceedings of the International Conference on Intelligent Computing and Communication Techniques ({ICICCT} 2024), New Delhi, India, 28--29 June, 2024 (Volume 2)},publisher={CRC Press},year={2025},month=jun,pages={240--246},isbn={978-1-032-86965-0},}
Elevating Aerial Vision-Language Navigation with Enhanced Feature Extraction Backbone Using GELAN and PGIIn Intelligent Computing and Communication Techniques: Proceedings of the International Conference on Intelligent Computing and Communication Techniques (ICICCT 2024), New Delhi, India, 28–29 June, 2024 (Volume 2) , Jun 2025