When VRP encoder is DINOv2, how to define mid-level and high-level features?
suyan451 opened this issue · 1 comments
suyan451 commented
When VRP encoder is DINOv2, how to define mid-level and high-level features?
syp2ysy commented
self.layer1, self.layer2, self.layer3, self.layer4 = nn.Sequential(*self.dinov2.blocks[:8]), nn.Sequential(*self.dinov2.blocks[8:12]), nn.Sequential(*self.dinov2.blocks[12:21]), nn.Sequential(*self.dinov2.blocks[21:24])