Fida Mohammad Thoker
Scholar

Fida Mohammad Thoker

Google Scholar ID: vqN8M3MAAAAJ
University of Amsterdam
Computer VisionAction RecognitionDeep Learning
Homepage↗Google Scholar↗
Citations & Impact
All-time
Citations
351
 
H-index
5
 
i10-index
5
 
Publications
12
 
Co-authors
0
 
Contact
CVOpen ↗TwitterOpen ↗GitHubOpen ↗LinkedInOpen ↗
Publications
7 items
Composition, Not Conversation: VLMs Lose the Scene, Not the Thread
2026
Cited
0
MedCTA: A Benchmark for Clinical Tool Agents
2026
Cited
0
FLORO: A Multimodal Geospatial Foundation Model for Ecological Remote Sensing Across Sensors and Scales
2026
Cited
0
TrackMAE: Video Representation Learning via Track Mask and Predict
2026
Cited
0
SEVERE++: Evaluating Benchmark Sensitivity in Generalization of Video Representation Learning
2025
Cited
0
SMILE: Infusing Spatial and Motion Semantics in Masked Video Learning
2025
Cited
0
Structured-Noise Masked Modeling for Video, Audio and Beyond
2025
Cited
0
Resume
Academic Achievements
  • Publications: Includes but not limited to SMILE: Infusing Spatial and Motion Semantics in Masked Video Learning (CVPR 2025), LocoMotion: Learning Motion-Focused Video-Language Representations (ACCV 2024 Oral), SIGMA: Sinkhorn-Guided Masked Video Modeling (ECCV 2024) etc.
  • Talks and Participation: Such as SkatingVerse Workshop, ICCV 2023 Doctoral Consortium, International Computer Vision Summer School (ICVSS) 2022 etc.
Research Experience
  • March 2024 - Present: Postdoctoral Researcher at KAUST;
  • May 2019 - December 2023: PhD Candidate at VIS LAB, University of Amsterdam.
Education
  • PhD: University of Amsterdam, advised by Prof. Cees Snoek;
  • Currently a Postdoctoral Researcher at King Abdullah University of Science and Technology (KAUST), working with Prof. Bernard Ghanem.
Background
  • Research Interests: Video Understanding, training foundation models via self-supervised learning from multiple modalities of the video data. Completed PhD at the University of Amsterdam, focusing on Video-Efficient Foundation Models.