Ankita SHUKLA

Responsable : Clément Chion

Mots clés : Visual wildlife monitoring, fine-grained recognition, robust models for small data, geometric constraints

Titre du projet : Beluga Re-Identification for population monitoring

Projet de recherche :  Aid in the research and development of tools for individual identification of beluga whales. Individual Identification of beluga whales is a challenging task. What makes it harder as compared to other species like tigers and leopards is the absence of unique patterns on their bodies. Additionally, the limited data and skewed data distribution across different identities makes it challenging to employ off the shelf deep learning solution. Therefore, in this project we aim to pull tools from traditional machine learning, geometry and deep learning to achieve our desired task.  

Communications scientifiques : 

Publications : 

1. Shukla, Ankita, et al. "Primate face identification in the wild." Pacific Rim International Conference on Artificial Intelligence. Springer, Cham, 2019.

2.  Shukla, Ankita, et al. "A hybrid approach to tiger re-identification." Proceedings of the IEEE/CVF International Conference on Computer Vision Workshops. 2019.

Diplômes et cursus universitaire : 

PhD from IIIT-Delhi India 

Currently working as a Postdoctoral researcher at Arizona State University