Nikolay
Nikolay
Hello Why we cant merge (gen_merged_model.py) all batchNorm on training -- all BN are constant batch_norm_param {use_global_stats: true} and some Scale on first 2 blocks (until conv3 block) Additional questions:...
Hello I have strange results when trying to visualize training process: [fast_rcnn/train.py](https://github.com/dereyly/pva-faster-rcnn/blob/master/lib/fast_rcnn/train.py) ``` rois = net.blobs['rois'].data bbox_targets = net.blobs['bbox_targets'].data boxes = rois[:, 1:5] tr_boxes = bbox_transform_inv(boxes, bbox_targets) tr_boxes = clip_boxes(tr_boxes,...
Hello. Thank you for nice work, I try to use non local nets (GCNet) on practice This config DCN + GCNet r4 + scale_augmentation and without mask -- faster RCNN...
Hello Maybe not right place to ask What best library to train detector? I have expirience in train FaterRCNN (and PVA), SSD in Caffe, but Caffe is too old, and...
Hello. Nice project! good stats of resnet38a: resnet38a | 20.66/5.27 | ../.. | 19.25/4.66 | ../.. It is your own model? I saw it in detection too, is it correct:...
Thank you for your great work! Do you try GIoU or CIoU loss in your experiments?
Hello I try to reproduce MTCNNv1 WIDER score on validation data I have not exactly same resault as in paper or mat files eval/plot/baselines/Val/setting_int/multitask-cascade-cnn/wider_pr_info_multitask-cascade-cnn_easy_val.mat I set parameters: threshold=[0.5 0.5 0.3];...
How to run model on (one of task) inference I found that in demo.py only features space Im interesting in human parsing task