Anima Majumder
Papers
4
Total Citations
20
H-Index
3
About
Anima Majumder is a researcher whose work bridges computer vision and robotics, with a focus on automation for e-commerce and warehouse environments. Her key research areas include monocular depth estimation, robotic manipulation, and deep learning-based annotation. She made significant contributions by developing an unsupervised deep learning framework that uses Bayesian inference to predict per-pixel depth and confidence maps from single RGB images, enhancing the reliability of 3D scene understanding. Her work on the automated Robotic Pick & Stow System for e-commerce warehouses, which integrates perception, localization, and manipulation modules, directly addresses real-world logistics challenges. Majumder also pioneered deep network-based automatic annotation techniques to reduce manual labor in generating training datasets for cluttered warehouse scenes, a problem highlighted by the Amazon Robotics Challenge. With over 20 citations across her most-cited papers, her research demonstrates practical impact in industrial automation. Notably, her annotation framework tackles the dual challenges of reducing manual effort and improving segmentation accuracy in densely packed environments, showcasing her ability to translate complex AI methods into scalable solutions for retail and logistics.
Research Focus
Key Achievements
Top Papers
- 1
- 2
- 3Deep Network based Automatic Annotation for Warehouse Automation4 citations · 2018
- 4