def forward(self, x):
x1 = self.bb(x)
p6 = self.p6(x1[-1])
p7 = self.p7(p6)
outs = x1 + [p6, p7]
sources = [time_to_batch(item)[0] for item in outs]
return sources
def _embeddings_loss(self, emb, ids, cls_targets, batchsize):
"""
In Time, promote similarity between targets with same id
:param pred_embeddings: [TN, #anchors, D]
:param ids: [TN, #nanchors]
:return:
"""
pos = (cls_targets > 0).view(-1)
emb = emb.view(-1, emb.size(-1))[pos]
ids = batch_to_time(ids, batchsize)
offsets = 100 * torch.arange(batchsize)[None, :, None].to(ids.device)
ids = ids + offsets
ids = time_to_batch(ids)[0]
ids = ids.view(-1)[pos]
#TODO:
#1. do this per frame not globally (benchmark this)
#2. weight id with iou loss = iou * id * (1-cos_mat) + (1-id) * max(0, cos_mat - margin)
# We do cos(x1, x2) between every vectors
# When they have same id = we use loss as 1 - cos (penalize for being far)
# When they have not same id = max(0, cos - margin) (penalize for not being close)
y = (ids[:, None] == ids[None, :])
norms = torch.norm(emb, dim=1)
norm_matrix = norms[:, None] * norms[None, :]
dot_matrix = torch.mm(emb, emb.t())
cos_matrix = dot_matrix / norm_matrix
margin = 0.5
loss = torch.where(y, 1 - cos_matrix, F.relu(cos_matrix - margin))
return loss.mean()
def forward(self, x):
x = self.conv1(x)
outs = []
for conv in self.conv2:
x = conv(x)
outs.append(time_to_batch(x)[0])
return outs
def forward(self, x):
sources = list()
x2 = self.conv1(x)
x3 = self.conv3(x2)
x4 = self.conv4(x3)
x5 = self.conv5(x4)
x6 = self.conv6(x5)
sources = [time_to_batch(item)[0] for item in [x3, x4, x5, x6]]
return sources
def upsample(self, x, feat_shape):
x, n = time_to_batch(x)
x = F.interpolate(x, size=feat_shape, mode='nearest')
x = batch_to_time(x, n)
return x
def forward_parallel(self, x):
t, n = x.size(0), x.size(1)
x = time_to_batch(x)[0]
x = self.module(x)
x = batch_to_time(x, n)
return x
def sequence_upsample(x, y):
x, n = time_to_batch(x)
x = F.interpolate(x, size=y.shape[-2:], mode='nearest')
x = batch_to_time(x, n)
return x
def forward(self, x):
x1 = self.conv1(x)
outs = self.conv2(x1)
outs = [time_to_batch(item)[0] for item in outs]
return outs
def forward(self, x):
x1 = self.bb(x)
outs = self.neck(x1)
sources = [time_to_batch(item)[0] for item in outs]
return sources
def forward(self, x):
x1 = self.conv1(x)
outs = self.conv2(x1)[-self.levels:]
sources = [time_to_batch(item)[0] for item in outs][::-1]
return sources