search

sanghoon / pva-faster-rcnn

Demo code for PVANet
https://arxiv.org/abs/1611.08588
Other
651 stars 241 forks source link

PVA+RFCN #53

Open KeyKy opened 7 years ago

KeyKy commented 7 years ago

I read your paper and merge your base PVANet with RFCN. Have you tried using PVANet with RFCN? I found it hard to train and the accuracy(ignore_label=0) gets improved slowly. I use voc_2007_trainval+voc_2012_trainval for training, voc_2007_test for testing and the final AP=0.69. The detection time is 0.070s faster than PVANet+Faster-RCNN. Could you give me some tips on it? Thanks!

xfguo-ucas commented 7 years ago

@KeyKy hi, I want to merge the base PVANet with SSD. But I don't know which pre-trained model should be use. Can you give me a tip? Thanks!

chengshuai commented 7 years ago

Hi @KeyKy @Morde-kaiser

I try train the example_train_384 in pvanet_obsolete(pva9.0) for pascal_voc 2007(trian:2007+2012,test:2007), the mAP is so low 13.1%，however, the mAP is 82.8% when i use sanghoon's model. Why the result is different? Is there other change or trick during your training?

The below is the script:

Training for 100k iterations

tools/train_net.py --gpu 0 --solver models/pvanet_obsolete/example_train_384/solver.prototxt --weights models/pvanet_obsolete/imagenet/original.model --iters 100000 --cfg models/pvanet_obsolete/cfgs/train.yml --imdb voc_2007_trainval Testing

tools/test_net.py --gpu 0 --def models/pvanet_obsolete/example_train_384/test.prototxt --net output/faster_rcnn_pvanet/voc_2007_trainval/pvanet_frcnn_384_iter_100000.caffemodel --cfg models/pvanet_obsolete/cfgs/submit_160715.yml

Do you retrain the model , and what is the mAP? Why the mAP is so low? How do merge the PVA + R-FCN and PVA+ SSD , can you give me a tips?

Thanks