What we're about

The Haifa branch of Datahack is meant to facilitate learning and discussion on various topics in data science, machine learning and statistics, with focus on the people and the companies working and doing research in Haifa and the area.

We aim to do these using several initiatives, starting with the Haifa flavor of the DataTalks meetup series, a series for co-learning and other future initiatives.

Join our Facebook group!
https://www.facebook.com/groups/datahackhaifa

Upcoming events (1)

DataTalks HFA #10 @ Elbit

Elbit Systems Ltd

We are excited to have Elbit host our 10th meetup, and the first around the fast-pacing topic of computer vision! Hadas Kogan, director of Elbit's department of artificial intelligence and computer vision, will share about her group's work on using cutting edge technology and sensor fusion to improve flight operations. Noa Alkobi, a computer vision researcher, will present her recent work on internal diverse image completion.

❗ IMPORTANT NOTE❗
Elbit's security requires attendees to list their ID number and bring an ID document with them. We kindly ask you to fill this form in order to participate. Also, registration will close on Tuesday January 31st. Community members that will not have registered in the above form by 31/01 will not be able to enter the Elbit premises.

♦ Time: February 5th, 17:30
♦ Location: Elbit, MATAM
♦ Language: The talks will be given in Hebrew
♦ Background: Basic knowledge in data science and machine learning is advised

------------------------------------------------------------------------------------------------

Abstracts for the talks:
AI In The Sky - Hadas Kogan (Elbit)
Real-time processing of radar, LiDAR, and camera data mounted on a helicopter for safety applications involves the use of advanced sensor technology and artificial intelligence algorithms. The sensors collect and process data of the surrounding environment in all weather conditions. The processing pipeline includes separation of ground and objects, identification and classification of obstacles, reconstruction of the terrain, and creating a coherent and clear semantic representation of the surrounding to the pilot.
Data from the different sensors is collected using Elbit’s flying lab, and is used to train semantic segmentation and detection neural networks. The high-resolution and high-accuracy information collected by the LiDAR is used as labelled data to train an encoder-decoder radar-based objects detection NN.
Overall, the integration of AI with advanced sensor technology can greatly enhance the safety and efficiency of flight operations.

Internal Diverse Image Completion - Noa Alkobi (Technion)
Image completion is widely used in photo restoration and editing applications, e.g. for object removal. Recently, there has been a surge of research on generating diverse completions for missing regions. However, existing methods require large training sets from a specific domain of interest, and often fail on general-content images. In this paper, we propose a diverse completion method that does not require a training set and can thus treat arbitrary images from any domain. Our internal diverse completion (IDC) approach draws inspiration from recent single-image generative models that are trained on multiple scales of a single image, adapting them to the extreme setting in which only a small portion of the image is available for training. We illustrate the strength of IDC on several datasets, using both user studies and quantitative comparisons.

Past events (19)

DataTalks HFA #9 @ Yahoo!

Andrei Sakharov St 9

Find us also at