Scholar
Favyen Bastani
Google Scholar ID: BTZBaosAAAAJ
MIT CSAIL
Citations & Impact
Citations
1,881
H-index
21
i10-index
24
Publications
20
Co-authors
0
Publications
Resume
Academic Achievements
- OTIF: Efficient Tracker Pre-processing over Large Video Datasets (SIGMOD 2022)
- Self-Supervised Multi-Object Tracking with Cross-Input Consistency (NeurIPS 2021)
- SkyQuery: An Aerial Drone Video Sensing Platform (Onward! 2021)
- Updating Street Maps using Changes Detected in Satellite Imagery (SIGSPATIAL 2021)
- Beyond Road Extraction: A Dataset for Map Update using Aerial Images (ICCV 2021)
- Vaas: Video Analytics at Scale (VLDB 2020, Demonstration)
- MIRIS: Fast Object Track Queries in Video (SIGMOD 2020)
- Sat2Graph: Road Graph Extraction through Graph-Tensor Encoding (ECCV 2020)
Background
- Currently an applied research scientist at the Allen Institute for AI.
- PhD student in Computer Science at the Massachusetts Institute of Technology (MIT), working with Sam Madden in the Data Systems group.
- Research focuses on lowering the barrier to using machine learning for video analytics, enabling broader and more socially beneficial applications.
- Interested in privacy-preserving video analytics, reducing labeling effort via self-supervision, and optimizing ML pipeline efficiency.
- Explores compelling applications in ecology, traffic safety, urban planning, autonomous robotics, and media analysis.