Papers
1
Total Citations
5
H-Index
1
About
Fabio Carrara is a leading researcher in computer vision and machine learning, with a focus on open-world perception, fine-grained visual understanding, and the limitations of large-scale vision-language models. His work critically examines the bottlenecks in modern AI systems, particularly the challenges posed by models like CLIP when applied to tasks requiring nuanced, fine-grained recognition in dynamic, real-world environments. Carrara’s most-cited paper, "Is CLIP the main roadblock for fine-grained open-world perception?" (2024, 5 citations), highlights his ability to identify and articulate key obstacles in deploying vision models for emerging domains such as extended reality, robotics, and autonomous driving. By probing the boundaries of open-world stimuli, he has contributed to advancing the flexibility and adaptability of computer vision systems, enabling them to handle novel concepts not encountered during training. His research is pivotal for bridging the gap between static, closed-set benchmarks and the unpredictable, open-world demands of modern applications. Carrara’s work continues to shape the future of perception systems, making him a notable voice in the quest for truly adaptive and robust AI.
Research Focus
Key Achievements
Top Papers
- 1Is CLIP the main roadblock for fine-grained open-world perception?5 citations · 2024