Upload folder using huggingface_hub
Browse files- config.json +141 -0
- configuration.json +1 -0
- model.safetensors +3 -0
config.json
ADDED
|
@@ -0,0 +1,141 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"cano_pose_type": 1,
|
| 3 |
+
"deform_head_type": "attn",
|
| 4 |
+
"dense_sample_pts": 160000,
|
| 5 |
+
"neural_renderer": {
|
| 6 |
+
"depth": 4,
|
| 7 |
+
"enc_channels": 128,
|
| 8 |
+
"enc_patch_size": 7,
|
| 9 |
+
"gradient_checkpointing": true,
|
| 10 |
+
"img_dim": 1024,
|
| 11 |
+
"type": "patch_4dptonly"
|
| 12 |
+
},
|
| 13 |
+
"neural_renderer_input": "feats",
|
| 14 |
+
"neural_renderer_patch_size": 1,
|
| 15 |
+
"encoder_feat_dim": 1024,
|
| 16 |
+
"encoder_freeze": true,
|
| 17 |
+
"encoder_grad_ckpt": true,
|
| 18 |
+
"encoder_model_name": "dinov2_vitl14_reg",
|
| 19 |
+
"encoder_type": "dinov2",
|
| 20 |
+
"expr_param_dim": 100,
|
| 21 |
+
"facesr": true,
|
| 22 |
+
"fix_opacity": false,
|
| 23 |
+
"fix_rotation": false,
|
| 24 |
+
"gs_clip_scaling": [
|
| 25 |
+
0,
|
| 26 |
+
0.05,
|
| 27 |
+
0.05,
|
| 28 |
+
3000
|
| 29 |
+
],
|
| 30 |
+
"gs_mlp_network_config": {
|
| 31 |
+
"activation": "silu",
|
| 32 |
+
"n_hidden_layers": 2,
|
| 33 |
+
"n_neurons": 512
|
| 34 |
+
},
|
| 35 |
+
"gs_query_dim": 128,
|
| 36 |
+
"gs_rendering": "featbacksplat",
|
| 37 |
+
"gs_sh": 3,
|
| 38 |
+
"gs_use_rgb": true,
|
| 39 |
+
"gs_xyz_offset_max_step": 1.0,
|
| 40 |
+
"human_model_path": "./pretrained_models/human_model_files",
|
| 41 |
+
"latent_query_points_type": "e2e_points",
|
| 42 |
+
"model_name": "LHMA4O",
|
| 43 |
+
"pcl_dim": 1024,
|
| 44 |
+
"render_features": true,
|
| 45 |
+
"shape_param_dim": 10,
|
| 46 |
+
"smplx_subdivide_num": 1,
|
| 47 |
+
"smplx_type": "smplx_diffused_voxel",
|
| 48 |
+
"tf_grad_ckpt": true,
|
| 49 |
+
"transformer_decoder": {
|
| 50 |
+
"freeze_image": true,
|
| 51 |
+
"freeze_point": true,
|
| 52 |
+
"image_backbone": {
|
| 53 |
+
"aa_order": [
|
| 54 |
+
"frame"
|
| 55 |
+
],
|
| 56 |
+
"depth": 3
|
| 57 |
+
},
|
| 58 |
+
"merge_ratio": 0.5,
|
| 59 |
+
"point_backbone": {
|
| 60 |
+
"attn_drop": 0.0,
|
| 61 |
+
"dec_channels": [
|
| 62 |
+
128,
|
| 63 |
+
256,
|
| 64 |
+
512
|
| 65 |
+
],
|
| 66 |
+
"dec_depths": [
|
| 67 |
+
4,
|
| 68 |
+
4,
|
| 69 |
+
4
|
| 70 |
+
],
|
| 71 |
+
"dec_num_head": [
|
| 72 |
+
16,
|
| 73 |
+
16,
|
| 74 |
+
16
|
| 75 |
+
],
|
| 76 |
+
"dec_patch_size": [
|
| 77 |
+
4096,
|
| 78 |
+
4096,
|
| 79 |
+
2048
|
| 80 |
+
],
|
| 81 |
+
"decoder_concat_stop_feat": 2,
|
| 82 |
+
"drop_path": 0.0,
|
| 83 |
+
"enable_flash": true,
|
| 84 |
+
"enable_rpe": false,
|
| 85 |
+
"enc_channels": [
|
| 86 |
+
64,
|
| 87 |
+
128,
|
| 88 |
+
256,
|
| 89 |
+
512
|
| 90 |
+
],
|
| 91 |
+
"enc_depths": [
|
| 92 |
+
4,
|
| 93 |
+
4,
|
| 94 |
+
4,
|
| 95 |
+
4
|
| 96 |
+
],
|
| 97 |
+
"enc_mode": false,
|
| 98 |
+
"enc_num_head": [
|
| 99 |
+
4,
|
| 100 |
+
8,
|
| 101 |
+
16,
|
| 102 |
+
16
|
| 103 |
+
],
|
| 104 |
+
"enc_patch_size": [
|
| 105 |
+
4096,
|
| 106 |
+
4096,
|
| 107 |
+
2048,
|
| 108 |
+
1024
|
| 109 |
+
],
|
| 110 |
+
"freeze_encoder": false,
|
| 111 |
+
"in_channels": 6,
|
| 112 |
+
"mask_token": false,
|
| 113 |
+
"mlp_ratio": 4,
|
| 114 |
+
"order": [
|
| 115 |
+
"z",
|
| 116 |
+
"z-trans",
|
| 117 |
+
"hilbert",
|
| 118 |
+
"hilbert-trans"
|
| 119 |
+
],
|
| 120 |
+
"pre_norm": true,
|
| 121 |
+
"proj_drop": 0.0,
|
| 122 |
+
"qk_scale": null,
|
| 123 |
+
"qkv_bias": true,
|
| 124 |
+
"shuffle_orders": false,
|
| 125 |
+
"stride": [
|
| 126 |
+
1,
|
| 127 |
+
2,
|
| 128 |
+
2
|
| 129 |
+
],
|
| 130 |
+
"traceable": true,
|
| 131 |
+
"upcast_attention": false,
|
| 132 |
+
"upcast_softmax": false
|
| 133 |
+
},
|
| 134 |
+
"type": "patch_efficient_pvt_mm_encoder_decoder_dense"
|
| 135 |
+
},
|
| 136 |
+
"transformer_dim": 1024,
|
| 137 |
+
"transformer_heads": 16,
|
| 138 |
+
"transformer_layers": 1,
|
| 139 |
+
"transformer_type": "mm",
|
| 140 |
+
"use_face_id": true
|
| 141 |
+
}
|
configuration.json
ADDED
|
@@ -0,0 +1 @@
|
|
|
|
|
|
|
| 1 |
+
{"framework":"Pytorch","task":"image-to-3D"}
|
model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:1897beb3c1ab8f0e198892f9813870aa82d2ada7ec4d043bb8cb293e1d65b725
|
| 3 |
+
size 4527981124
|