MTMT
MTMT copied to clipboard
Code for the CVPR 2020 paper "A Multi-task Mean Teacher for Semi-supervised Shadow Detection"
error
img, target, edge = self.joint_transform(img, target, edge) TypeError: __call__() takes 3 positional arguments but 4 were given
I use the trained model you provided in BaiduNetDisk (iter_10000.pth) ![image](https://user-images.githubusercontent.com/19933627/98695838-0def4b80-23ae-11eb-92c0-a5008abfe578.png) It seems your model doesn't match the checkpoint, so that the following error occurs: ``` RuntimeError: Error(s) in loading...
in my opinion,the mt just make the teacher and student network's outputs as the same as possible,but the same result doesn't mean the result is a correct result,especially when the...
代码问题 ema
# build the whole network def build_model(base_model_cfg='resnext101', ema=False): if not ema: return TUN_bone(base_model_cfg, *extra_layer(base_model_cfg, ResNeXt101())) else: return TUN_bone(base_model_cfg, *extra_layer(base_model_cfg, ResNeXt101())) 怎么感觉不管这里ema的值是什么,返回的都是一个内容呢?
数据集问题
你好!请问哪里可以下载UCF数据集?找了很久没找到..
I couldn't find UCF datasets. The link provided authors only have 110 images, but the original UCF dataset have 245 images. Who can help me? If you have the link,...
Hi, would it be possible to upload the trained model on google drive, like you did with the detection results? I can't create a baidu account without a chinese phone...
Thanks for your work. I can detect the shadow from an image now, the output of the network is a binary map, but how to remove the shadow region from...
def __init__(self, primary_indices, secondary_indices, batch_size, secondary_batch_size): self.primary_indices = primary_indices self.secondary_indices = secondary_indices self.secondary_batch_size = secondary_batch_size self.primary_batch_size = batch_size - secondary_batch_size #print(len(self.secondary_indices)) #print(len(self.primary_indices)) #print(self.primary_batch_size) #print(self.secondary_batch_size) assert len(self.primary_indices) >= self.primary_batch_size > 0...
从论文的Table 2 来看,引入MT后确实给detection 任务带来了提升,但是不太能理解的一点就是给teacher network输入unlabeled data 后,其实这部分的loss计算是来自于unlabeled送入student net来获取的一个伪label,这里的unlabeled 数据没有真实的GT参考,为什么加上M最后可以给basic net带来增益和提升的呢,针对这一点,师兄是怎么理解的呢