preTrained predictions are ready for observation

maskcut/dino/dinoUsed.py

@@ -0,0 +1,346 @@
+# Copyright (c) Meta Platforms, Inc. and affiliates.
+
+"""
+Copied from Dino repo. https://github.com/facebookresearch/dino
+Mostly copy-paste from timm library.
+https://github.com/rwightman/pytorch-image-models/blob/master/timm/models/vision_transformer.py
+"""
+import math
+from functools import partial
+
+import torch
+import torch.nn as nn
+
+def _no_grad_trunc_normal_(tensor, mean, std, a, b):
+    # Cut & paste from PyTorch official master until it's in a few official releases - RW
+    # Method based on https://people.sc.fsu.edu/~jburkardt/presentations/truncated_normal.pdf
+    def norm_cdf(x):
+        # Computes standard normal cumulative distribution function
+        return (1. + math.erf(x / math.sqrt(2.))) / 2.
+
+    if (mean < a - 2 * std) or (mean > b + 2 * std):
+        warnings.warn("mean is more than 2 std from [a, b] in nn.init.trunc_normal_. "
+                      "The distribution of values may be incorrect.",
+                      stacklevel=2)
+
+    with torch.no_grad():
+        # Values are generated by using a truncated uniform distribution and
+        # then using the inverse CDF for the normal distribution.
+        # Get upper and lower cdf values
+        l = norm_cdf((a - mean) / std)
+        u = norm_cdf((b - mean) / std)
+
+        # Uniformly fill tensor with values from [l, u], then translate to
+        # [2l-1, 2u-1].
+        tensor.uniform_(2 * l - 1, 2 * u - 1)
+
+        # Use inverse cdf transform for normal distribution to get truncated
+        # standard normal
+        tensor.erfinv_()
+
+        # Transform to proper mean, std
+        tensor.mul_(std * math.sqrt(2.))
+        tensor.add_(mean)
+
+        # Clamp to ensure it's in the proper range
+        tensor.clamp_(min=a, max=b)
+        return tensor
+
+
+def trunc_normal_(tensor, mean=0., std=1., a=-2., b=2.):
+    # type: (Tensor, float, float, float, float) -> Tensor
+    return _no_grad_trunc_normal_(tensor, mean, std, a, b)
+
+
+def drop_path(x, drop_prob: float = 0., training: bool = False):
+    if drop_prob == 0. or not training:
+        return x
+    keep_prob = 1 - drop_prob
+    shape = (x.shape[0],) + (1,) * (x.ndim - 1)  # work with diff dim tensors, not just 2D ConvNets
+    random_tensor = keep_prob + torch.rand(shape, dtype=x.dtype, device=x.device)
+    random_tensor.floor_()  # binarize
+    output = x.div(keep_prob) * random_tensor
+    return output
+
+
+class DropPath(nn.Module):
+    """Drop paths (Stochastic Depth) per sample  (when applied in main path of residual blocks).
+    """
+    def __init__(self, drop_prob=None):
+        super(DropPath, self).__init__()
+        self.drop_prob = drop_prob
+
+    def forward(self, x):
+        return drop_path(x, self.drop_prob, self.training)
+
+
+class Mlp(nn.Module):
+    def __init__(self, in_features, hidden_features=None, out_features=None, act_layer=nn.GELU, drop=0.):
+        super().__init__()
+        out_features = out_features or in_features
+        hidden_features = hidden_features or in_features
+        self.fc1 = nn.Linear(in_features, hidden_features)
+        self.act = act_layer()
+        self.fc2 = nn.Linear(hidden_features, out_features)
+        self.drop = nn.Dropout(drop)
+
+    def forward(self, x):
+        x = self.fc1(x)
+        x = self.act(x)
+        x = self.drop(x)
+        x = self.fc2(x)
+        x = self.drop(x)
+        return x
+
+
+class Attention(nn.Module):
+    def __init__(self, dim, num_heads=8, qkv_bias=False, qk_scale=None, attn_drop=0., proj_drop=0.):
+        super().__init__()
+        self.num_heads = num_heads
+        head_dim = dim // num_heads
+        self.scale = qk_scale or head_dim ** -0.5
+
+        self.qkv = nn.Linear(dim, dim * 3, bias=qkv_bias)
+        self.attn_drop = nn.Dropout(attn_drop)
+        self.proj = nn.Linear(dim, dim)
+        self.proj_drop = nn.Dropout(proj_drop)
+
+    def forward(self, x):
+        B, N, C = x.shape
+        qkv = self.qkv(x).reshape(B, N, 3, self.num_heads, C // self.num_heads).permute(2, 0, 3, 1, 4)
+        q, k, v = qkv[0], qkv[1], qkv[2]
+
+        attn = (q @ k.transpose(-2, -1)) * self.scale
+        attn = attn.softmax(dim=-1)
+        attn = self.attn_drop(attn)
+
+        x = (attn @ v).transpose(1, 2).reshape(B, N, C)
+        x = self.proj(x)
+        x = self.proj_drop(x)
+        return x, attn
+
+
+class Block(nn.Module):
+    def __init__(self, dim, num_heads, mlp_ratio=4., qkv_bias=False, qk_scale=None, drop=0., attn_drop=0.,
+                 drop_path=0., act_layer=nn.GELU, norm_layer=nn.LayerNorm):
+        super().__init__()
+        self.norm1 = norm_layer(dim)
+        self.attn = Attention(
+            dim, num_heads=num_heads, qkv_bias=qkv_bias, qk_scale=qk_scale, attn_drop=attn_drop, proj_drop=drop)
+        self.drop_path = DropPath(drop_path) if drop_path > 0. else nn.Identity()
+        self.norm2 = norm_layer(dim)
+        mlp_hidden_dim = int(dim * mlp_ratio)
+        self.mlp = Mlp(in_features=dim, hidden_features=mlp_hidden_dim, act_layer=act_layer, drop=drop)
+
+    def forward(self, x, return_attention=False):
+        y, attn = self.attn(self.norm1(x))
+        if return_attention:
+            return attn
+        x = x + self.drop_path(y)
+        x = x + self.drop_path(self.mlp(self.norm2(x)))
+        return x
+
+
+class PatchEmbed(nn.Module):
+    """ Image to Patch Embedding
+    """
+    def __init__(self, img_size=224, patch_size=16, in_chans=3, embed_dim=768):
+        super().__init__()
+        num_patches = (img_size // patch_size) * (img_size // patch_size)
+        self.img_size = img_size
+        self.patch_size = patch_size
+        self.num_patches = num_patches
+
+        self.proj = nn.Conv2d(in_chans, embed_dim, kernel_size=patch_size, stride=patch_size)
+
+    def forward(self, x):
+        B, C, H, W = x.shape
+        x = self.proj(x).flatten(2).transpose(1, 2)
+        return x
+
+
+class VisionTransformer(nn.Module):
+    """ Vision Transformer """
+    def __init__(self, img_size=[224], patch_size=16, in_chans=3, num_classes=0, embed_dim=768, depth=12,
+                 num_heads=12, mlp_ratio=4., qkv_bias=False, qk_scale=None, drop_rate=0., attn_drop_rate=0.,
+                 drop_path_rate=0., norm_layer=nn.LayerNorm, **kwargs):
+        super().__init__()
+        self.num_features = self.embed_dim = embed_dim
+
+        self.patch_embed = PatchEmbed(
+            img_size=img_size[0], patch_size=patch_size, in_chans=in_chans, embed_dim=embed_dim)
+        num_patches = self.patch_embed.num_patches
+
+        self.cls_token = nn.Parameter(torch.zeros(1, 1, embed_dim))
+        self.pos_embed = nn.Parameter(torch.zeros(1, num_patches + 1, embed_dim))
+        self.pos_drop = nn.Dropout(p=drop_rate)
+
+        dpr = [x.item() for x in torch.linspace(0, drop_path_rate, depth)]  # stochastic depth decay rule
+        self.blocks = nn.ModuleList([
+            Block(
+                dim=embed_dim, num_heads=num_heads, mlp_ratio=mlp_ratio, qkv_bias=qkv_bias, qk_scale=qk_scale,
+                drop=drop_rate, attn_drop=attn_drop_rate, drop_path=dpr[i], norm_layer=norm_layer)
+            for i in range(depth)])
+        self.norm = norm_layer(embed_dim)
+
+        # Classifier head
+        self.head = nn.Linear(embed_dim, num_classes) if num_classes > 0 else nn.Identity()
+
+        trunc_normal_(self.pos_embed, std=.02)
+        trunc_normal_(self.cls_token, std=.02)
+        self.apply(self._init_weights)
+
+    def _init_weights(self, m):
+        if isinstance(m, nn.Linear):
+            trunc_normal_(m.weight, std=.02)
+            if isinstance(m, nn.Linear) and m.bias is not None:
+                nn.init.constant_(m.bias, 0)
+        elif isinstance(m, nn.LayerNorm):
+            nn.init.constant_(m.bias, 0)
+            nn.init.constant_(m.weight, 1.0)
+
+    def interpolate_pos_encoding(self, x, w, h):
+        npatch = x.shape[1] - 1
+        N = self.pos_embed.shape[1] - 1
+        if npatch == N and w == h:
+            return self.pos_embed
+        class_pos_embed = self.pos_embed[:, 0]
+        patch_pos_embed = self.pos_embed[:, 1:]
+        dim = x.shape[-1]
+        w0 = w // self.patch_embed.patch_size
+        h0 = h // self.patch_embed.patch_size
+        # we add a small number to avoid floating point error in the interpolation
+        # see discussion at https://github.com/facebookresearch/dino/issues/8
+        w0, h0 = w0 + 0.1, h0 + 0.1
+        patch_pos_embed = nn.functional.interpolate(
+            patch_pos_embed.reshape(1, int(math.sqrt(N)), int(math.sqrt(N)), dim).permute(0, 3, 1, 2),
+            scale_factor=(w0 / math.sqrt(N), h0 / math.sqrt(N)),
+            mode='bicubic',
+        )
+        assert int(w0) == patch_pos_embed.shape[-2] and int(h0) == patch_pos_embed.shape[-1]
+        patch_pos_embed = patch_pos_embed.permute(0, 2, 3, 1).view(1, -1, dim)
+        return torch.cat((class_pos_embed.unsqueeze(0), patch_pos_embed), dim=1)
+
+    def prepare_tokens(self, x):
+        B, nc, w, h = x.shape
+        x = self.patch_embed(x)  # patch linear embedding
+
+        # add the [CLS] token to the embed patch tokens
+        cls_tokens = self.cls_token.expand(B, -1, -1)
+        x = torch.cat((cls_tokens, x), dim=1)
+
+        # add positional encoding to each token
+        x = x + self.interpolate_pos_encoding(x, w, h)
+
+        return self.pos_drop(x)
+
+    def forward(self, x):
+        x = self.prepare_tokens(x)
+        for blk in self.blocks:
+            x = blk(x)
+        x = self.norm(x)
+        return x[:, 0]
+
+    def get_last_selfattention(self, x):
+        x = self.prepare_tokens(x)
+        for i, blk in enumerate(self.blocks):
+            if i < len(self.blocks) - 1:
+                x = blk(x)
+            else:
+                # return attention of the last block
+                return blk(x, return_attention=True)
+
+    def get_intermediate_layers(self, x, n=1):
+        x = self.prepare_tokens(x)
+        # we return the output tokens from the `n` last blocks
+        output = []
+        for i, blk in enumerate(self.blocks):
+            x = blk(x)
+            if len(self.blocks) - i <= n:
+                output.append(self.norm(x))
+        return output
+
+
+def vit_small(patch_size=16, **kwargs):
+    model = VisionTransformer(
+        patch_size=patch_size, embed_dim=384, depth=12, num_heads=6, mlp_ratio=4,
+        qkv_bias=True, norm_layer=partial(nn.LayerNorm, eps=1e-6), **kwargs)
+    return model
+
+
+def vit_base(patch_size=16, **kwargs):
+    model = VisionTransformer(
+        patch_size=patch_size, embed_dim=768, depth=12, num_heads=12, mlp_ratio=4,
+        qkv_bias=True, norm_layer=partial(nn.LayerNorm, eps=1e-6), **kwargs)
+    return model
+
+class ViTFeat(nn.Module):
+    """ Vision Transformer """
+    def __init__(self, pretrained_pth, feat_dim, vit_arch = 'base', vit_feat = 'k', patch_size=16):
+        super().__init__()
+        if vit_arch == 'base' :
+            self.model = vit_base(patch_size=patch_size, num_classes=0)
+
+        else :
+            self.model = vit_small(patch_size=patch_size, num_classes=0)
+
+        self.feat_dim = feat_dim
+        self.vit_feat = vit_feat
+        self.patch_size = patch_size
+
+#        state_dict = torch.load(pretrained_pth, map_location="cpu")
+        state_dict = torch.hub.load_state_dict_from_url(pretrained_pth)
+        self.model.load_state_dict(state_dict, strict=True)
+        print('Loading weight from {}'.format(pretrained_pth))
+
+
+    def forward(self, img) :
+        feat_out = {}
+        def hook_fn_forward_qkv(module, input, output):
+            feat_out["qkv"] = output
+
+        self.model._modules["blocks"][-1]._modules["attn"]._modules["qkv"].register_forward_hook(hook_fn_forward_qkv)
+
+
+        # Forward pass in the model
+        with torch.no_grad() :
+            h, w = img.shape[2], img.shape[3]
+            feat_h, feat_w = h // self.patch_size, w // self.patch_size
+            attentions = self.model.get_last_selfattention(img)
+            bs, nb_head, nb_token = attentions.shape[0], attentions.shape[1], attentions.shape[2]
+            qkv = (
+                    feat_out["qkv"]
+                    .reshape(bs, nb_token, 3, nb_head, -1)
+                    .permute(2, 0, 3, 1, 4)
+                )
+            q, k, v = qkv[0], qkv[1], qkv[2]
+
+            k = k.transpose(1, 2).reshape(bs, nb_token, -1)
+            q = q.transpose(1, 2).reshape(bs, nb_token, -1)
+            v = v.transpose(1, 2).reshape(bs, nb_token, -1)
+
+            # Modality selection
+            if self.vit_feat == "k":
+                feats = k[:, 1:].transpose(1, 2).reshape(bs, self.feat_dim, feat_h * feat_w)
+            elif self.vit_feat == "q":
+                feats = q[:, 1:].transpose(1, 2).reshape(bs, self.feat_dim, feat_h * feat_w)
+            elif self.vit_feat == "v":
+                feats = v[:, 1:].transpose(1, 2).reshape(bs, self.feat_dim, feat_h * feat_w)
+            elif self.vit_feat == "kqv":
+                k = k[:, 1:].transpose(1, 2).reshape(bs, self.feat_dim, feat_h * feat_w)
+                q = q[:, 1:].transpose(1, 2).reshape(bs, self.feat_dim, feat_h * feat_w)
+                v = v[:, 1:].transpose(1, 2).reshape(bs, self.feat_dim, feat_h * feat_w)
+                feats = torch.cat([k, q, v], dim=1)
+            return feats
+
+
+if __name__ == "__main__":
+    vit_arch = 'base'
+    vit_feat = 'k'
+
+    model = ViTFeat(vit_arch, vit_feat)
+    img = torch.cuda.FloatTensor(4, 3, 224, 224)
+    model.cuda()
+    # Forward pass in the model
+    feat = model(img)
+    print (feat.shape)

maskcut/maskcut.py

@@ -24,8 +24,8 @@ import dino
 # modfied by Xudong Wang based on third_party/TokenCut
 sys.path.append('../')
 sys.path.append('../third_party')
-from token_cut.unsupervised_saliency_detection import utils, metric
-from token_cut.unsupervised_saliency_detection.object_discovery import detect_box
+from unsupervised_saliency_detection import utils, metric
+from unsupervised_saliency_detection.object_discovery import detect_box
 # bilateral_solver codes are modfied based on https://github.com/poolio/bilateral_solver/blob/master/notebooks/bilateral_solver.ipynb
 # from TokenCut.unsupervised_saliency_detection.bilateral_solver import BilateralSolver, BilateralGrid
 # crf codes are are modfied based on https://github.com/lucasb-eyer/pydensecrf/blob/master/pydensecrf/tests/test_dcrf.py