Hugging Face's logo

facebook
/
sapiens2-matting-1b

Image Segmentation
sapiens
sapiens2
human-centric
matting
Model card Files
xet
 Community
1

Add HF-compatible model.safetensors, config.json, and preprocessor_config.json

#1
by guarin HF Staff - opened
base: refs/heads/main
←
from: refs/pr/1
Discussion Files changed
+195
-0
Files changed (3) hide show
  1. config.json +168 -0
  2. model.safetensors +3 -0
  3. preprocessor_config.json +24 -0
config.json ADDED
@@ -0,0 +1,168 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "apply_layernorm": true,
3
+ "architectures": [
4
+ "Sapiens2ForMatting"
5
+ ],
6
+ "attention_dropout": 0.0,
7
+ "drop_path_rate": 0.0,
8
+ "head_conv_kernel_sizes": [
9
+ 3,
10
+ 3,
11
+ 3
12
+ ],
13
+ "head_conv_out_channels": [
14
+ 64,
15
+ 32,
16
+ 16
17
+ ],
18
+ "head_scale_conv_kernel_sizes": null,
19
+ "head_scale_conv_out_channels": null,
20
+ "head_scale_final_hidden_sizes": null,
21
+ "head_upsample_kernel_sizes": [
22
+ 3,
23
+ 3,
24
+ 3,
25
+ 3
26
+ ],
27
+ "head_upsample_out_channels": [
28
+ 768,
29
+ 512,
30
+ 256,
31
+ 128
32
+ ],
33
+ "hidden_act": "silu",
34
+ "hidden_size": 1536,
35
+ "id2label": {
36
+ "0": "LABEL_0",
37
+ "1": "LABEL_1",
38
+ "2": "LABEL_2",
39
+ "3": "LABEL_3"
40
+ },
41
+ "image_size": [
42
+ 1024,
43
+ 768
44
+ ],
45
+ "initializer_range": 0.02,
46
+ "intermediate_size": 6144,
47
+ "key_bias": true,
48
+ "label2id": {
49
+ "LABEL_0": 0,
50
+ "LABEL_1": 1,
51
+ "LABEL_2": 2,
52
+ "LABEL_3": 3
53
+ },
54
+ "layer_norm_eps": 1e-06,
55
+ "layerscale_value": 1.0,
56
+ "mlp_bias": true,
57
+ "model_type": "sapiens2",
58
+ "num_attention_heads": 24,
59
+ "num_channels": 3,
60
+ "num_hidden_layers": 40,
61
+ "num_key_value_heads_per_layer": [
62
+ 24,
63
+ 24,
64
+ 24,
65
+ 24,
66
+ 24,
67
+ 24,
68
+ 24,
69
+ 24,
70
+ 12,
71
+ 12,
72
+ 12,
73
+ 12,
74
+ 12,
75
+ 12,
76
+ 12,
77
+ 12,
78
+ 12,
79
+ 12,
80
+ 12,
81
+ 12,
82
+ 12,
83
+ 12,
84
+ 12,
85
+ 12,
86
+ 12,
87
+ 12,
88
+ 12,
89
+ 12,
90
+ 12,
91
+ 12,
92
+ 12,
93
+ 12,
94
+ 24,
95
+ 24,
96
+ 24,
97
+ 24,
98
+ 24,
99
+ 24,
100
+ 24,
101
+ 24
102
+ ],
103
+ "num_register_tokens": 8,
104
+ "out_features": [
105
+ "stage40"
106
+ ],
107
+ "out_indices": [
108
+ 40
109
+ ],
110
+ "patch_size": 16,
111
+ "pos_embed_dtype": "bfloat16",
112
+ "pos_embed_jitter": null,
113
+ "pos_embed_rescale": 2.0,
114
+ "pos_embed_shift": null,
115
+ "proj_bias": true,
116
+ "query_bias": true,
117
+ "reshape_hidden_states": true,
118
+ "rope_theta": 100.0,
119
+ "semantic_loss_ignore_index": 255,
120
+ "stage_names": [
121
+ "stem",
122
+ "stage1",
123
+ "stage2",
124
+ "stage3",
125
+ "stage4",
126
+ "stage5",
127
+ "stage6",
128
+ "stage7",
129
+ "stage8",
130
+ "stage9",
131
+ "stage10",
132
+ "stage11",
133
+ "stage12",
134
+ "stage13",
135
+ "stage14",
136
+ "stage15",
137
+ "stage16",
138
+ "stage17",
139
+ "stage18",
140
+ "stage19",
141
+ "stage20",
142
+ "stage21",
143
+ "stage22",
144
+ "stage23",
145
+ "stage24",
146
+ "stage25",
147
+ "stage26",
148
+ "stage27",
149
+ "stage28",
150
+ "stage29",
151
+ "stage30",
152
+ "stage31",
153
+ "stage32",
154
+ "stage33",
155
+ "stage34",
156
+ "stage35",
157
+ "stage36",
158
+ "stage37",
159
+ "stage38",
160
+ "stage39",
161
+ "stage40"
162
+ ],
163
+ "transformers_version": "5.8.0.dev0",
164
+ "use_gated_mlp": true,
165
+ "use_mask_token": false,
166
+ "use_qk_norm": true,
167
+ "value_bias": true
168
+ }
model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2a0ea4ea2234e8dbd51e952950d8cedfe3ee2eca9099003784fd2d9cec6bc6e1
3
+ size 6157412312
preprocessor_config.json ADDED
@@ -0,0 +1,24 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "do_normalize": true,
3
+ "do_pad": false,
4
+ "do_reduce_labels": false,
5
+ "do_rescale": true,
6
+ "do_resize": true,
7
+ "image_mean": [
8
+ 0.485,
9
+ 0.456,
10
+ 0.406
11
+ ],
12
+ "image_processor_type": "Sapiens2ImageProcessor",
13
+ "image_std": [
14
+ 0.229,
15
+ 0.224,
16
+ 0.225
17
+ ],
18
+ "resample": 2,
19
+ "rescale_factor": 0.00392156862745098,
20
+ "size": {
21
+ "height": 1024,
22
+ "width": 768
23
+ }
24
+ }