Deep Learning for Robust Normal Estimation in Unstructured Point Clouds
Normal estimation in point clouds is a crucial first step for numerous algorithms, from surface reconstruction and scene understanding to
Read more
Luigi Freda
Computer Vision, AI, Robotics Engineer, PhD
Posts
Hierarchical Progressive Multiview Stereo
Progressive multiview stero is an implementation of an algorihm taking a sparse 3D model (result of structure from motion software)
Read more
SpotMini: A New Creature from Boston Dynamics
SpotMini is a new smaller version of the Spot robot. It can be equipped with an arm and is all-electric (no hydraulics).
Read more
High-Performance and Tunable Stereo Reconstruction
In this work, Researchers of MIT present an high-performance and tunable stereo disparity estimation method. It succeeds to reach a
Read more
Snotbot: Whale Research with Aerial Technology
This video presents the beautiful work of the Snotbot’s team of researchers on their Sea of Cortez expedition. Aerial technology is adapted
Read more
Robotic Laundry
How about asking your robot to make a complete laundry cycle?…PR2 seems almost there. See more on this page.
Read more
Long Short-Term Memory Networks (LSTMs)
In this nice post, LSTMs are described to have more control than typical RNNs over what they remember. In general,
Read more
History of Neural Network Design
From Eugenio Curlurciello’s blog, Professor at Purdue University, you can find here a post about the the history of neural network design
Read more
Deep Reinforcement Learning: from Pong Pixels to a Learned Player
Here you can find a fantastic post on reinforcement learning by Andrej Karpathy. The Author explains how it is possible
Read more
Nanowire-based Battery Material that can be Recharged Hundreds of Thousands of Times
University of California, Irvine researchers have invented nanowire-based battery material that can be recharged hundreds of thousands of times. This
Read more
Probabilistic Volumetric Reconstruction using Ray Potentials
From the webpage of the project “This paper presents a novel probabilistic foundation for volumetric 3-d reconstruction. We formulate the problem as
Read more
Active Volumetric 3D Reconstruction
A new open-source library has been released by the Robotics and Perception Group of Davide Scaramuzza. Technical details of the
Read more
Itseez Acquired by Intel
OpenCV originated at Intel, and is now back at Intel. People wonder about the possibility that OpenCV can stop beeing
Read more
Autonomous Mini Rally Car Teaches Itself to Powerslide
In this work from Georgia Tech, researchers have developed a control algorithm that allows small-scale autonomous cars to drive autonomously around
Read more