Benchmark RadarVietFashion
VietFashion is a benchmark for sketch-text composed image retrieval centered on the Ao Dai, a traditional Vietnamese garment. It includes 650 sketches expanded to over 21,000 photorealistic images with captions, and adopts a multi-target retrieval setting.
- Released
- 2026-06-11
- Readiness
- Inspectable
- Primary field
- General AI
Why it matters
Cultural garments require fine-grained retrieval systems that understand subtle structural and symbolic details. VietFashion exposes gaps in modeling cultural semantics and multi-modal composition.
Motivation
Cultural garments pose a unique challenge for visual retrieval systems, as their identity often depends on subtle structural and symbolic details that are poorly captured by standard AI models.
Primary resources
Benchmark Radar records only publicly supported details and links back to primary sources for verification.