Scholar
Amir Habibian
Google Scholar ID: RZ9pOY4AAAAJ
Research Scientist, Qualcomm AI Research
Citations & Impact
Citations
1,131
H-index
17
i10-index
22
Publications
20
Co-authors
97
list available
Publications
Browse publications on Google Scholar (top-right) ↗
Resume (English only)
Academic Achievements
- - Published multiple papers on topics such as video diffusion models, mobile video editing, and spatially aware object insertion.
- - Demonstrated first-in-industry diffusion-based video editing running on a phone at NeurIPS 2024.
- - Showcased using diffusion models to generate training and test data for perception models in Automotive at CES 2024.
Research Experience
- - Led teams at Qualcomm developing mobile architectures for image and video generation, video perception, and neural video compression.
- - Organized the eLVM: Efficient Large Vision Models workshop at CVPR 2024, targeting the computational efficiency of training and inference for foundation models.
- - Co-organized the Resource Efficient Deep Learning for Computer Vision workshop at ICCV 2023, discussing recent progress on the computational efficiency of computer vision models.
Education
- PhD from the University of Amsterdam, focusing on learning multimodal video representations.
Background
- Interested in the computational efficiency of neural networks. Works as a research scientist at Qualcomm AI Research.
Miscellany
- Involved in several talks and presentations on efficient generation of visual content.
