Diff between caffe models.

[logic1988]

What's the difference between four caffe models? Which model has the highest accuracy?Which one equal to the pytorch/pwc_net.pth.tar? Thank you!

[deqings]

Sorry for the confusion and late reply.

./model/pwc_net.caffemodel PWC-Net_ROB fine-tuned using Sintel, KITTI, and HD1K data; tends to perform robustly across different datasets

./model/pwc_net_chairs_things.caffemodel PWC-Net-ROB pre-trained using FlyingChairs and FlyingThings, probably lowest accruacy

./model/PWC-Net_plus/PWC-Net_plus_sintel/pwc-net-plus-sintel.caffemodel PWC-Net+ fined-tuned on Sintel, as reported in Table 1 of this extended journal version, highest accuracy on Sintel benchmark

./model/PWC-Net_plus/PWC-Net_plus_kitti/pwc-net-plus-kitti.caffemodel PWC-Net+ fined-tuned on KITTI, as reported in this Table 3 extended journal version, highest accuracy on KITTI benchmark