Robust Deepfake Detection, NTIRE 2026 Challenge: Report

TL;DR

The NTIRE 2026 Challenge focused on developing deepfake detectors robust to image degradations, emphasizing real-world applicability and resilience against malicious manipulations.

Contribution

This report introduces a challenge addressing robustness in deepfake detection, highlighting methods that utilize foundation models, ensembles, and degradation training.

Findings

Top methods used large foundation models and ensembles.

Degradation training improved robustness against image distortions.

Limited test data exposure prevented overfitting.

Abstract

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 image degradation. In addition to weaker degradations that can accidentally occur in the image processing pipeline, there is another risk of malicious deepfakes that specifically introduce degradations, purposefully exploiting the detector's weaknesses in that regard. Here, we present an overview of the NTIRE 2026 Robust Deepfake Detection Challenge, which specifically addresses that problem. Participants were tasked with building a detector that would later be tested on an unknown test-set, which included both common and uncommon degradations of various strengths. With a total number of 337 participants and 57 submissions to the final leaderboard, the first edition of the challenge was well received. To ensure the reliability of the results, participants were given only 24h to complete the test run with no labels provided, limiting the possibility of training on the test data. Furthermore, the top solutions were scored on a private test-set to detect any such overfitting. This report presents the competition setting, dataset preparation, as well as details and performance of methods. Top methods rely on large foundation models, ensembles, and degradation training to combine generality and robustness.