The VQualA 2026 Challenge aims to advance reliable, interpretable, and human- and machine-aligned visual quality intelligence for next-generation AI systems. Supported by the IEEE Signal Processing Society Challenge Program, it is open to researchers, students, and practitioners who are interested in visual quality assessment, multimodal learning, large vision-language models, and related areas.

The challenge includes two tracks:

• MPD: Machine-oriented Image Quality Assessment Challenge – focusing on how visual distortions affect machine perception and downstream intelligent tasks, with an emphasis on machine-oriented quality prediction and task-aware perceptual evaluation.
• LongVQU-Eval: Long-Term Video Quality Understanding Evaluation – evaluating the ability of large vision-language models to understand, reason about, and explain perceptual quality degradations across long-duration videos.

Teams may participate through either of the two competition tracks, subject to the challenge rules and submission requirements.

The challenge offers a total prize pool of USD 5,000.

Key dates

• Registration deadline: October 15, 2026
• Test set release: October 15, 2026
• Test-set results submission deadline: October 30, 2026
• Methodology report and open-source model weights due: November 4, 2026
• Challenge results announcement: November 20, 2026

Click here to learn more.