Elad Hirsch

I am a PhD candidate at the Technion Institute of Technology in Haifa, Israel, under the guidance of Prof. Ayellet Tal from the Electrical & Computer Engineering faculty (CGM lab).

I am interested in visual perception and image understanding. My research focuses on vision-language tasks with limited data, specifically describing images with text in scenarios where paired data is scarce or unavailable. Hence, it involves visual understanding, language modeling, and multi-modal representation learning. Additionally, I have been teaching the 'Image Processing & Analysis' course for five years, where I am happy to share my enthusiasm for digital image understanding.

Email    Google Scholar    LinkedIn    GitHub

Elad Hirsch

News

[Sep 2024] Our new evaluation metric for medical report generation (soon on ArXiv) will be presented at NeurIPS 2024!
[July 2024] Our paper MedRAT was accepted to ECCV 2024
[Mar 2024] Our paper MedCycle was accepted to Findings of NAACL 2024
[Jan 2024] I will present our work on Asymmetric Image Retrieval using Ensembles in the first oral session of WACV
[Oct 2023] Two of our works, CLID and Asymmetric Image Retrieval using Ensembles, were accepted to WACV
[July 2023] Our work LIMITR was accepted to ICCV
[Jan 2023] I am mentoring the 10th-grade students of Kogel High School in their AI projects
[June 2022] I am interning at Meta Reality Labs for the summer, collaborating with Erez Farhan
[May 2022] I was awarded the Porat Prize for outstanding graduate students

Publications

Image-aware Evaluation of Generated Medical Reports
Gefen Dawidowicz*, Elad Hirsch*, Ayellet Tal
NeurIPS, 2024
(Soon on ArXiv)
MedRAT: Unpaired Medical Report Generation via Auxiliary Tasks
Elad Hirsch, Gefen Dawidowicz, Ayellet Tal
ECCV, 2024
paper / code
MedCycle: Unpaired Medical Report Generation via Cycle-Consistency
Elad Hirsch, Gefen Dawidowicz, Ayellet Tal
Findings of NAACL, 2024
paper / code
Asymmetric Image Retrieval with Cross Model Compatible Ensembles
Alon Shoshan*, Ori Linial*, Nadav Bhonker, Elad Hirsch, Lior Zamir, Igor Kviatkovsky,
Gérard Medioni
WACV, 2024 (Oral Presentation)
paper

Work done as part of a summer internship at Amazon.

CLID: Controlled-Length Image Descriptions with Limited Data
Elad Hirsch, Ayellet Tal
WACV, 2024
paper / code
LIMITR: Leveraging Local Information for Medical Image-Text Representation
Gefen Dawidowicz, Elad Hirsch, Ayellet Tal
ICCV, 2023
paper / code
Color Visual Illusions: A Statistics-based Computational Model
Elad Hirsch, Ayellet Tal
NeurIPS, 2020
paper / code

This website uses Jon Barron's template (source code). Big thanks to Jon for sharing the code!