Places365 CNNs

Convolutional neural networks (CNNs) trained on the Places2 Database can be used for scene recognition as well as generic deep scene features for visual recognition. We share the following pre-trained CNNs using Caffe and PyTorch.

We are actively working on the version 2 of Places Database. If you need to use the legacy dataset (the original Places365 or Places205) urgently for research purposes, please sign this form, thank you.