Antonino Furnari

About

Antonino Furnari is a computer vision researcher whose work spans fisheye image processing, action anticipation, egocentric vision, and embodied robot navigation. He first gained recognition for his foundational contributions to wide-angle camera processing, developing affine covariant features for fisheye distortion modeling (27 citations) and generalized Sobel filters for gradient estimation in distorted images (14 citations) — work that addressed critical challenges in automotive, surveillance, and robotics imaging systems. Furnari's research has since expanded into anticipatory intelligence, exploring how machines can predict human actions and object interactions from visual data. His knowledge distillation approach for action anticipation via label smoothing (10 citations) and survey on egocentric future prediction reflect a sustained commitment to enabling robots and assistive systems to understand and foresee human behavior. His contributions to affordance-based short-term interaction anticipation and attention modeling further demonstrate his versatility in this domain. More recently, Furnari has tackled the sim-to-real gap in robot visual navigation, benchmarking mid-level representations and evaluating embodied navigation through platforms like Habitat. His editorial role in machine vision for assistive technologies underscores his broader influence in shaping research directions. Altogether, his work bridges fundamental image analysis with applied intelligent systems, making him a notable contributor to human-centered computer vision.

Research Focus

Key Achievements

Top Papers

1. 1
   Affine Covariant Features for Fisheye Distortion Local Modeling

   27 citations · 2016
2. 2
   Generalized Sobel Filters for gradient estimation of distorted images

   14 citations · 2015
3. 3
   Knowledge Distillation for Action Anticipation via Label Smoothing

   10 citations · 2021
4. 4
   Image-based Navigation in Real-World Environments via Multiple Mid-level Representations: Fusion Models, Benchmark and Efficient Evaluation

   5 citations · 2023
5. 5
   AFF-ttention! Affordances and Attention Models for Short-Term Object Interaction Anticipation

   5 citations · 2024
6. 6
   Predicting the future from first person (egocentric) vision: A survey

   4 citations · 2021
7. 7
   A Survey on Human-aware Robot Navigation

   4 citations · 2021
8. 8
   On Embodied Visual Navigation in Real Environments Through Habitat

   2 citations · 2021
9. 9
   Editorial: Machine Vision for Assistive Technologies

   2 citations · 2022

Key Collaborators

Contact & Links