Teaching AI to Smell: Meet the New York Smells Dataset
Computers can see and hear, but smelling the world has been out of reach. New research introduces New York Smells: a large, real-world dataset that pairs odors with images.
- 7,000 smell-image pairs from 3,500 objects, indoors and outdoors — about 70x more objects than past datasets.
- Benchmark tasks: match a smell to its photo, recognize scenes/objects/materials from smell alone, and even tell apart different grass species.
- Results: visual data helps AI learn better smell representations, and these learned features beat widely used hand-crafted ones.
Why it matters: Smell-aware AI could help robots detect hazards, check food freshness, or understand environments more like living creatures do.
Paper: https://arxiv.org/abs/2511.20544v1
Paper: https://arxiv.org/abs/2511.20544v1
Register: https://www.AiFeta.com
AI MachineLearning ComputerVision Olfaction Dataset Robotics Research Multimodal
