Places365
by MIT CSAIL · free · Last verified 2026-03-17
Places365 is a scene-centric database with 1.8 million training images across 365 scene categories, designed to train and evaluate scene recognition models. The dataset enables models to understand the semantic meaning of places and environments, making it ideal for applications in autonomous driving, robotics, and image retrieval.
http://places2.csail.mit.edu ↗B+
B+—Good
Adoption: AQuality: AFreshness: C+Citations: AEngagement: F
Specifications
- License
- CC-BY (non-commercial variants available)
- Pricing
- free
- Capabilities
- scene-classification, transfer-learning, place-recognition
- Integrations
- PyTorch, TensorFlow, HuggingFace Datasets
- Use Cases
- model-training, benchmark, scene-understanding
- API Available
- No
- Tags
- scene-recognition, scene-classification, transfer-learning, indoor-outdoor
- Added
- 2026-03-17
- Completeness
- 100%
Index Score
70.7Adoption
80
Quality
87
Freshness
50
Citations
85
Engagement
0