27 papers
NTIRE 2026 Low-light Enhancement: Twilight Cowboy Challenge
Aleksei Khalin, Egor Ershov, Artyom Panshin +46
This paper presents a review of the NTIRE 2026 Low-light Enhancement: Twilight Cowboy Challenge. The objective of the competition was to merge a set of misaligned smartphone images…
Learning the Arabic Dialect Continuum as a Continuous Space: A Regression Approach to Speaker Origin Prediction
Mohamed Aziz Khadraoui, Adel Ammar, Bilel Benjdira +3
We present a regression-based approach to Arabic dialect geolocation that models dialectal variation as a continuous geographic space rather than discrete categories. Speaker origi…
Low Light Image Enhancement Challenge at NTIRE 2026
George Ciubotariu, Sharif S M A, Abdur Rehman +90
This paper presents a comprehensive review of the NTIRE 2026 Low Light Image Enhancement Challenge, highlighting the proposed solutions and final results. The objective of this cha…
NTIRE 2026 The Second Challenge on Day and Night Raindrop Removal for Dual-Focused Images: Methods and Results
Xin Li, Yeying Jin, Suhang Yao +95
This paper presents an overview of the NTIRE 2026 Second Challenge on Day and Night Raindrop Removal for Dual-Focused Images. Building upon the success of the first edition, this c…
Robust Deepfake Detection, NTIRE 2026 Challenge: Report
Benedikt Hopf, Radu Timofte, Chenfan Qu +54
Robustness is a long-overlooked problem in deepfake detection. However, detection performance is nearly worthless in the real world if it suffers under exposure to even slight imag…
Trust-SSL: Additive-Residual Selective Invariance for Robust Aerial Self-Supervised Learning
Wadii Boulila, Adel Ammar, Bilel Benjdira +1
Self-supervised learning (SSL) is a standard approach for representation learning in aerial imagery. Existing methods enforce invariance between augmented views, which works well w…