Home Page of Helio Perroni Filho
About Me
"Helio was an 8-year-old boy who got hold of a computer and never let go." Programming was my first foreign language, English being a distant second; today it feels like second nature. After getting my B.Sc. in Computer Science I spent the next ten years or so working in enterprise IT, before returning to academia for research in robotics and artificial vision. Currently I'm working in the robotics industry, developing mobile robot navigation software.
Profile
- 2022: Senior Robotics Engineer, Elder Laboratory of Human and Computer Vision
- 2018: Principal Software Architect, Autonomous Robotics at CrossWing Inc.
- 2017 (Nov): Post-doc in autonomous driving technologies, Laboratório de Computação de Alto Desempenho (LCAD)
- 2017 (Oct): Self-Driving Car Engineer nanodegree, Udacity
- 2016: Ph.D. in Computer Science, University of Tsukuba
- 2013 (Oct): Ph.D. student at the Intelligent Robot Laboratory
- 2013 (Apr): Admitted to Brazilian scholarship program "Ciência sem Fonteiras" ("Science without Borders")
- 2010: M.Sc. in Computer Science, UFES
- 2008: Software Engineer at LG Electronics do Brasil
- 2004: Software developer at MOGAI Tecnologia de Informação
- 2004: B.Sc. in Computer Science, UFES
Research Interests
- Autonomous navigation, especially teach / replay
- Cognitive architectures (e.g. Semantic Pointer Architecture, IDEAL)
- Computer vision
- Alternative neural network architectures (e.g. Holographic Reduced Representation, Hierarchical Temporal Memory)
Publications
- Attentive Sensing for Long-Range Face Recognition. IEEE/CVF Winter Conference on Applications of Computer Vision (WACV) Long-Range Recognition (LRR) Workshop 2023. ^
- Aleksander Trajcevski, Helio Perroni Filho et al. Sensorimotor System Design of Socially Intelligent Robots. 26th International Conference on Pattern Recognition, 2022.
- A Biologically Inspired Architecture for Visual Self-location. Biologically Inspired Cognitive Architectures (BICA) for Young Scientists, 2016, pp. 297-303. DOI: 10.1007/978-3-319-32554-5_38
- Image Correspondence Based on Interest Point Correlation in Difference Streams: Method and Applications to Mobile Robot Localization. Journal of Robotics & Mechatronics, Volume 28, No. 2, 2016, pp. 234-241. DOI: 10.20965/jrm.2016.p0234
- Mobile Robot Path Drift Estimation using Visual Streams. IEEE/SICE International Symposium on System Integration (SII), Tokyo, Japan, December 13-15, 2014, Proceedings, pp. 192-197. DOI: 10.1109/SII.2014.7028036
- On multichannel neurons, with an application to template search. Journal of Network and Innovative Computing, Volume 2, Issue 1, 2014, pp. 10-21. ISSN 2160-2174
- A biology-based template-matching framework. 12th International Conference Intelligent Systems Design and Applications (ISDA), Kochi, India, November 27-29, 2012, Proceedings, pp. 326-331. DOI: 10.1109/ISDA.2012.6416559
- VG-RAM WNN Approach to Monocular Depth Perception. 17th International Conference on Neural Information Processing (ICONIP), Sydney, Australia, November 22-25, 2010, Proceedings, Part II, pp.509-516. DOI: 10.1007/978-3-642-17534-3_63
Downloads
Favourite Links
- The Register - "Biting the hand that feeds IT". A news portal on technology, natural sciences, economy and law, (almost) always with a view to the IT field – and an occasional bit of tinkering and silliness on the side. Easily the sharpest wordsmiths in the IT reporting arena.
- Rewiring Neuroscience - Probably my favorite science book. Starting with a startling assumption – that contrary to conventional wisdom, individual neuron spikes can somehow transmit meaningful information from one neuron to the next – author John Harris brings together a number of overlooked experimental results into a surprisingly feasible theory of how the brain works.
- Atomic Insights - Among other indiscretions, I am an atomic power apologist. Rod Adams maintains my favorite blog on nuclear energy science and industry. Posts are sometimes dry, focused on legal and political issues, but they always give good insights on the challenges and perspectives of the field.
- The Gervais Principle - Venkatesh Rao's account of human nature, social order and organizational dynamics is a must-read, even if you've never watched The Office – I for one didn't, and still this is one of my all-time favorite books.