TickTockVQA: Analog Clock Reading Images for Vision-Language Models

12,483 images of analog clocks with time labels support training and evaluation of Vision-Language Models. The dataset originates from research presented at CVPR 2026 Findings. It was created by jaeha-choi and last updated on May 13, 2026.

Use Cases

• Training VQA models to read analog clock times based on image inputs.
• Evaluating model performance on clock-hand spatial reasoning tasks.
• Benchmarking Vision-Language Models on a specialized visual understanding challenge.

Strengths

• Contains 12,483 labeled images, providing a substantial corpus for model training.
• Specifically designed for the task of analog clock reading, offering a focused benchmark.

Limitations

• Column-level documentation is absent; field semantics must be inferred after download.
• The dataset's copyright and redistribution conditions are noted as belonging to the original data corpora, not the authors.
• Row count is unknown, which may limit suitability assessment.

Provenance

Source

Collected from publicly available data corpora.

Freshness

Last updated 2026-05-13 04:30:51