File size: 16,638 Bytes
83ae704 cc2404f 83ae704 7a6bc02 83ae704 cc2404f 83ae704 7a6bc02 83ae704 7a6bc02 83ae704 7a6bc02 83ae704 cc2404f f964355 cc2404f 83ae704 7a6bc02 83ae704 7a6bc02 83ae704 cc2404f 7a6bc02 83ae704 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 |
#!/usr/bin/env python3
# Copyright (C) 2024-present Naver Corporation. All rights reserved.
# Licensed under CC BY-NC-SA 4.0 (non-commercial use only).
#
# --------------------------------------------------------
# gradio demo
# --------------------------------------------------------
import math
import gradio
import os
import torch
import numpy as np
import tempfile
import functools
import trimesh
import copy
from scipy.spatial.transform import Rotation
from mast3r.cloud_opt.sparse_ga import sparse_global_alignment
from mast3r.cloud_opt.tsdf_optimizer import TSDFPostProcess
from mast3r.model import AsymmetricMASt3R
from mast3r.utils.misc import hash_md5
import mast3r.utils.path_to_dust3r # noqa
from dust3r.image_pairs import make_pairs
from dust3r.utils.image import load_images
from dust3r.utils.device import to_numpy
from dust3r.viz import add_scene_cam, CAM_COLORS, OPENGL, pts3d_to_trimesh, cat_meshes
from dust3r.demo import get_args_parser as dust3r_get_args_parser
import matplotlib.pyplot as pl
pl.ion()
torch.backends.cuda.matmul.allow_tf32 = True # for gpu >= Ampere and pytorch >= 1.12
batch_size = 1
def get_args_parser():
parser = dust3r_get_args_parser()
parser.add_argument('--share', action='store_true')
actions = parser._actions
for action in actions:
if action.dest == 'model_name':
action.choices = ["MASt3R_ViTLarge_BaseDecoder_512_catmlpdpt_metric"]
# change defaults
parser.prog = 'mast3r demo'
return parser
def _convert_scene_output_to_glb(outdir, imgs, pts3d, mask, focals, cams2world, cam_size=0.05,
cam_color=None, as_pointcloud=False,
transparent_cams=False, silent=False):
assert len(pts3d) == len(mask) <= len(imgs) <= len(cams2world) == len(focals)
pts3d = to_numpy(pts3d)
imgs = to_numpy(imgs)
focals = to_numpy(focals)
cams2world = to_numpy(cams2world)
scene = trimesh.Scene()
# full pointcloud
if as_pointcloud:
pts = np.concatenate([p[m.ravel()] for p, m in zip(pts3d, mask)])
col = np.concatenate([p[m] for p, m in zip(imgs, mask)])
pct = trimesh.PointCloud(pts.reshape(-1, 3), colors=col.reshape(-1, 3))
scene.add_geometry(pct)
else:
meshes = []
for i in range(len(imgs)):
meshes.append(pts3d_to_trimesh(imgs[i], pts3d[i].reshape(imgs[i].shape), mask[i]))
mesh = trimesh.Trimesh(**cat_meshes(meshes))
scene.add_geometry(mesh)
# add each camera
for i, pose_c2w in enumerate(cams2world):
if isinstance(cam_color, list):
camera_edge_color = cam_color[i]
else:
camera_edge_color = cam_color or CAM_COLORS[i % len(CAM_COLORS)]
add_scene_cam(scene, pose_c2w, camera_edge_color,
None if transparent_cams else imgs[i], focals[i],
imsize=imgs[i].shape[1::-1], screen_width=cam_size)
rot = np.eye(4)
rot[:3, :3] = Rotation.from_euler('y', np.deg2rad(180)).as_matrix()
scene.apply_transform(np.linalg.inv(cams2world[0] @ OPENGL @ rot))
outfile = os.path.join(outdir, 'scene.glb')
if not silent:
print('(exporting 3D scene to', outfile, ')')
scene.export(file_obj=outfile)
return outfile
def get_3D_model_from_scene(outdir, silent, scene, min_conf_thr=2, as_pointcloud=False, mask_sky=False,
clean_depth=False, transparent_cams=False, cam_size=0.05, TSDF_thresh=0):
"""
extract 3D_model (glb file) from a reconstructed scene
"""
if scene is None:
return None
# get optimized values from scene
rgbimg = scene.imgs
focals = scene.get_focals().cpu()
cams2world = scene.get_im_poses().cpu()
# 3D pointcloud from depthmap, poses and intrinsics
if TSDF_thresh > 0:
tsdf = TSDFPostProcess(scene, TSDF_thresh=TSDF_thresh)
pts3d, _, confs = to_numpy(tsdf.get_dense_pts3d(clean_depth=clean_depth))
else:
pts3d, _, confs = to_numpy(scene.get_dense_pts3d(clean_depth=clean_depth))
msk = to_numpy([c > min_conf_thr for c in confs])
return _convert_scene_output_to_glb(outdir, rgbimg, pts3d, msk, focals, cams2world, as_pointcloud=as_pointcloud,
transparent_cams=transparent_cams, cam_size=cam_size, silent=silent)
def get_reconstructed_scene(outdir, model, device, silent, image_size, filelist, optim_level, lr1, niter1, lr2, niter2,
min_conf_thr, matching_conf_thr, as_pointcloud, mask_sky, clean_depth, transparent_cams,
cam_size, scenegraph_type, winsize, win_cyclic, refid, TSDF_thresh, shared_intrinsics,
**kw):
"""
from a list of images, run mast3r inference, sparse global aligner.
then run get_3D_model_from_scene
"""
imgs = load_images(filelist, size=image_size, verbose=not silent)
if len(imgs) == 1:
imgs = [imgs[0], copy.deepcopy(imgs[0])]
imgs[1]['idx'] = 1
filelist = [filelist[0], filelist[0] + '_2']
scene_graph_params = [scenegraph_type]
if scenegraph_type in ["swin", "logwin"]:
scene_graph_params.append(str(winsize))
elif scenegraph_type == "oneref":
scene_graph_params.append(str(refid))
if scenegraph_type in ["swin", "logwin"] and not win_cyclic:
scene_graph_params.append('noncyclic')
scene_graph = '-'.join(scene_graph_params)
pairs = make_pairs(imgs, scene_graph=scene_graph, prefilter=None, symmetrize=True)
if optim_level == 'coarse':
niter2 = 0
# Sparse GA (forward mast3r -> matching -> 3D optim -> 2D refinement -> triangulation)
scene = sparse_global_alignment(filelist, pairs, os.path.join(outdir, 'cache'),
model, lr1=lr1, niter1=niter1, lr2=lr2, niter2=niter2, device=device,
opt_depth='depth' in optim_level, shared_intrinsics=shared_intrinsics,
matching_conf_thr=matching_conf_thr, **kw)
outfile = get_3D_model_from_scene(outdir, silent, scene, min_conf_thr, as_pointcloud, mask_sky,
clean_depth, transparent_cams, cam_size, TSDF_thresh)
return scene, outfile
def set_scenegraph_options(inputfiles, win_cyclic, refid, scenegraph_type):
num_files = len(inputfiles) if inputfiles is not None else 1
show_win_controls = scenegraph_type in ["swin", "logwin"]
show_winsize = scenegraph_type in ["swin", "logwin"]
show_cyclic = scenegraph_type in ["swin", "logwin"]
max_winsize, min_winsize = 1, 1
if scenegraph_type == "swin":
if win_cyclic:
max_winsize = max(1, math.ceil((num_files - 1) / 2))
else:
max_winsize = num_files - 1
elif scenegraph_type == "logwin":
if win_cyclic:
half_size = math.ceil((num_files - 1) / 2)
max_winsize = max(1, math.ceil(math.log(half_size, 2)))
else:
max_winsize = max(1, math.ceil(math.log(num_files, 2)))
winsize = gradio.Slider(label="Scene Graph: Window Size", value=max_winsize,
minimum=min_winsize, maximum=max_winsize, step=1, visible=show_winsize)
win_cyclic = gradio.Checkbox(value=win_cyclic, label="Cyclic sequence", visible=show_cyclic)
win_col = gradio.Column(visible=show_win_controls)
refid = gradio.Slider(label="Scene Graph: Id", value=0, minimum=0,
maximum=num_files - 1, step=1, visible=scenegraph_type == 'oneref')
return win_col, winsize, win_cyclic, refid
def main_demo(tmpdirname, model, device, image_size, server_name, server_port, silent=False, share=False):
if not silent:
print('Outputing stuff in', tmpdirname)
recon_fun = functools.partial(get_reconstructed_scene, tmpdirname, model, device, silent, image_size)
model_from_scene_fun = functools.partial(get_3D_model_from_scene, tmpdirname, silent)
with gradio.Blocks(css=""".gradio-container {margin: 0 !important; min-width: 100%};""", title="MASt3R Demo") as demo:
# scene state is save so that you can change conf_thr, cam_size... without rerunning the inference
scene = gradio.State(None)
gradio.HTML('<h2 style="text-align: center;">MASt3R Demo</h2>')
with gradio.Column():
inputfiles = gradio.File(file_count="multiple")
with gradio.Row():
with gradio.Column():
with gradio.Row():
lr1 = gradio.Slider(label="Coarse LR", value=0.07, minimum=0.01, maximum=0.2, step=0.01)
niter1 = gradio.Number(value=500, precision=0, minimum=0, maximum=10_000,
label="num_iterations", info="For coarse alignment!")
lr2 = gradio.Slider(label="Fine LR", value=0.014, minimum=0.005, maximum=0.05, step=0.001)
niter2 = gradio.Number(value=200, precision=0, minimum=0, maximum=100_000,
label="num_iterations", info="For refinement!")
optim_level = gradio.Dropdown(["coarse", "refine", "refine+depth"],
value='refine', label="OptLevel",
info="Optimization level")
with gradio.Row():
matching_conf_thr = gradio.Slider(label="Matching Confidence Thr", value=5.,
minimum=0., maximum=30., step=0.1,
info="Before Fallback to Regr3D!")
shared_intrinsics = gradio.Checkbox(value=False, label="Shared intrinsics",
info="Only optimize one set of intrinsics for all views")
scenegraph_type = gradio.Dropdown([("complete: all possible image pairs", "complete"),
("swin: sliding window", "swin"),
("logwin: sliding window with long range", "logwin"),
("oneref: match one image with all", "oneref")],
value='complete', label="Scenegraph",
info="Define how to make pairs",
interactive=True)
with gradio.Column(visible=False) as win_col:
winsize = gradio.Slider(label="Scene Graph: Window Size", value=1,
minimum=1, maximum=1, step=1)
win_cyclic = gradio.Checkbox(value=False, label="Cyclic sequence")
refid = gradio.Slider(label="Scene Graph: Id", value=0,
minimum=0, maximum=0, step=1, visible=False)
run_btn = gradio.Button("Run")
with gradio.Row():
# adjust the confidence threshold
min_conf_thr = gradio.Slider(label="min_conf_thr", value=1.5, minimum=0.0, maximum=10, step=0.1)
# adjust the camera size in the output pointcloud
cam_size = gradio.Slider(label="cam_size", value=0.2, minimum=0.001, maximum=1.0, step=0.001)
TSDF_thresh = gradio.Slider(label="TSDF Threshold", value=0., minimum=0., maximum=1., step=0.01)
with gradio.Row():
as_pointcloud = gradio.Checkbox(value=True, label="As pointcloud")
# two post process implemented
mask_sky = gradio.Checkbox(value=False, label="Mask sky")
clean_depth = gradio.Checkbox(value=True, label="Clean-up depthmaps")
transparent_cams = gradio.Checkbox(value=False, label="Transparent cameras")
outmodel = gradio.Model3D()
# events
scenegraph_type.change(set_scenegraph_options,
inputs=[inputfiles, win_cyclic, refid, scenegraph_type],
outputs=[win_col, winsize, win_cyclic, refid])
inputfiles.change(set_scenegraph_options,
inputs=[inputfiles, win_cyclic, refid, scenegraph_type],
outputs=[win_col, winsize, win_cyclic, refid])
win_cyclic.change(set_scenegraph_options,
inputs=[inputfiles, win_cyclic, refid, scenegraph_type],
outputs=[win_col, winsize, win_cyclic, refid])
run_btn.click(fn=recon_fun,
inputs=[inputfiles, optim_level, lr1, niter1, lr2, niter2, min_conf_thr, matching_conf_thr,
as_pointcloud, mask_sky, clean_depth, transparent_cams, cam_size,
scenegraph_type, winsize, win_cyclic, refid, TSDF_thresh, shared_intrinsics],
outputs=[scene, outmodel])
min_conf_thr.release(fn=model_from_scene_fun,
inputs=[scene, min_conf_thr, as_pointcloud, mask_sky,
clean_depth, transparent_cams, cam_size, TSDF_thresh],
outputs=outmodel)
cam_size.change(fn=model_from_scene_fun,
inputs=[scene, min_conf_thr, as_pointcloud, mask_sky,
clean_depth, transparent_cams, cam_size, TSDF_thresh],
outputs=outmodel)
TSDF_thresh.change(fn=model_from_scene_fun,
inputs=[scene, min_conf_thr, as_pointcloud, mask_sky,
clean_depth, transparent_cams, cam_size, TSDF_thresh],
outputs=outmodel)
as_pointcloud.change(fn=model_from_scene_fun,
inputs=[scene, min_conf_thr, as_pointcloud, mask_sky,
clean_depth, transparent_cams, cam_size, TSDF_thresh],
outputs=outmodel)
mask_sky.change(fn=model_from_scene_fun,
inputs=[scene, min_conf_thr, as_pointcloud, mask_sky,
clean_depth, transparent_cams, cam_size, TSDF_thresh],
outputs=outmodel)
clean_depth.change(fn=model_from_scene_fun,
inputs=[scene, min_conf_thr, as_pointcloud, mask_sky,
clean_depth, transparent_cams, cam_size, TSDF_thresh],
outputs=outmodel)
transparent_cams.change(model_from_scene_fun,
inputs=[scene, min_conf_thr, as_pointcloud, mask_sky,
clean_depth, transparent_cams, cam_size, TSDF_thresh],
outputs=outmodel)
demo.launch(share=False, server_name=server_name, server_port=server_port)
if __name__ == '__main__':
parser = get_args_parser()
args = parser.parse_args()
if args.server_name is not None:
server_name = args.server_name
else:
server_name = '0.0.0.0' if args.local_network else '127.0.0.1'
if args.weights is not None:
weights_path = args.weights
else:
weights_path = "naver/" + args.model_name
model = AsymmetricMASt3R.from_pretrained(weights_path).to(args.device)
chkpt_tag = hash_md5(weights_path)
# mast3r will write the 3D model inside tmpdirname/chkpt_tag
if args.tmp_dir is not None:
tmpdirname = args.tmp_dir
cache_path = os.path.join(tmpdirname, chkpt_tag)
os.makedirs(cache_path, exist_ok=True)
main_demo(cache_path, model, args.device, args.image_size, server_name, args.server_port, silent=args.silent,
share=args.share)
else:
with tempfile.TemporaryDirectory(suffix='_mast3r_gradio_demo') as tmpdirname:
cache_path = os.path.join(tmpdirname, chkpt_tag)
os.makedirs(cache_path, exist_ok=True)
main_demo(tmpdirname, model, args.device, args.image_size,
server_name, args.server_port, silent=args.silent,
share=args.share)
|