M. Hossein Daraei

PhD Candidate, Electrical Engineering
Jack Baskin School of Engineering
University of California, Santa Cruz

M. Hossein Daraei is a graduate student at Computer Vision Lab of University of California, Santa Cruz. His research area is Computer Vision, Signal Processing, and Sensor Fusion with applications to Autonomous Driving. He is also interested in Data Visualization, Machine Learning, and Creative Writing. He is pursuing a PhD in Electrical Engineering under the supervision of Roberto Manduchi. He is also collaborating with Volkswagen Group of America's Electronics Research Lab.

PhD Dissertation

Tightly-Coupled LiDAR And Camera For Autonomous Vehicles
Here is a visualization of my final measurement-level fusion algorithm for tracking obstacles -- with color hue encoding direction and color saturation encoding magnitude. Implemented on NVIDIA Jetson TK1, it runs at 1 frame/second -- processing the whole image and regions of different distances at different resolutions. The underlying data structures are 16x16 overlapping squares which are perfect for CUDA implementation. While each square is being tracked independently higher level connections between them are also taken into account. For further reading, please refer to my PhD dissertation.
phd thesis dense depth multilevel processing velocity estimation

ITSC 2017 Paper Highlight

Region Segmentation Using LiDAR and Camera

Inspired by the ideas behind superpixels, which segment an image into homogenous regions to accelerate subsequent processing steps (e.g. tracking), we present a sensor-fusion-based segmentation approach that generates dense depth regions referred to as supersurfaces. This method aggregates both a point cloud from a LiDAR and an image from a camera to provide an over-segmentation of the three-dimensional scene into piece-wise planar surfaces by utilizing a multi-label Markov Random Field (MRF). [read more]
project website paper poster slides

IV 2017 Paper Highlight

Velocity and Shape from Tightly-Coupled LiDAR and Camera

[3/15/17] In this paper, we propose a multi-object tracking and reconstruction approach through measurement-level fusion of LiDAR and camera. The proposed method, regardless of object class, estimates 3D motion and structure for all obstacles. Using an intermediate surface representation, measurements from both sensors are processed within a joint framework. We combine dense optical flow, surface reconstruction, and point-to-surface terms in a tightly-coupled non-linear energy function, which is minimized using Iterative Reweighted Least Squares (IRLS).
project website paper poster slides

US Patent Filed

[6/23/16] US Patent, 14/579,144   |  K. Raghu, M. Daraei, P. Natarajan, N. Lopez
Early detection of exit only and shared lanes using perception technology

An in-vehicle system for identifying exit-only lanes and shared exit lanes on a roadway having a first camera for obtaining image data regarding lane markings on the roadway, a second camera for obtaining image data regarding exit signs, a lane marking detection module for detecting lane markings on the roadway, an exit sign detection module for detecting exit signs, and an exit sign analyzer for detecting arrows [more] check patent website

Paper highlight

[10/25/14] "Optical Flow Problem in the Presence of Spatially-varying Motion Blur" has been accepted for presentation at International Symposium on Visual Computing (ISVC '14), in which Motion-Blur-aware Combined Local and Global method (MB-CLG) is presented for computing optical flow in videos.
project website paper code slides

Abstract--The problem of optical flow computation has various applications in Computer Vision, and serves as a key problem that has been well studied over the past decades. While most of the techniques for inferring optical flow are based on the brightness constancy assumption, various conditions including the presence of motion blur evidently violate this fundamental presumption. [continue reading]

Brain Drain From Iran

[12/13/2013] I have designed a data visualization on "Brain Drain From Iran", it specifically considers the statistics of bachelor graduates of Electrical Engineering department of Sharif University of Technology, demonstrating how they've scattered throughout the world and which cities and universities have been the most popular ones. This work is done as a course project for CMPS 263: Data Visualization course.

view high resolution version interview with take-part