File size: 14,169 Bytes
23e73f9 | 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 | import os
import sys
import random
import torch
import numpy as np
import torch.nn.functional as F
from argparse import ArgumentParser
from core.registry import register_method
from core.base_method import BaseMethod
sys.path.append(os.path.abspath(os.path.join(os.path.dirname(__file__), '../../GeoGaussian')))
from utils.loss_utils import l1_loss, ssim
from gaussian_renderer import render as native_render
from diff_gaussian_rasterization import SurfaceAlign
from scene import Scene, GaussianModel
from arguments import ModelParams, PipelineParams, OptimizationParams
@register_method("geogaussian")
class GeoGaussianWrapper(BaseMethod):
def __init__(self, dataset_config, hyperparams):
self.parser = ArgumentParser()
self.lp = ModelParams(self.parser)
self.op = OptimizationParams(self.parser)
self.pp = PipelineParams(self.parser)
self.args = self.parser.parse_args([])
self.args.source_path = dataset_config["source_path"]
self.args.model_path = dataset_config["model_path"]
self.args.eval = True
self.args.resolution = dataset_config.get("resolution", 1)
self.track_decoupling = hyperparams.get("track_decoupling", False)
self.dataset = self.lp.extract(self.args)
self.opt = self.op.extract(self.args)
self.pipe = self.pp.extract(self.args)
self.gaussians = GaussianModel(self.dataset.sh_degree)
# INJECTED_RES_FIX begin
import sys as _sys
_scene, _explicit_res = None, None
for _i, _a in enumerate(_sys.argv[:-1]):
_v = _sys.argv[_i + 1]
if _a == "--scene": _scene = _v
elif _a == "--source_path": _scene = _v.rstrip("/").split("/")[-1]
elif _a == "--resolution":
try: _explicit_res = int(_v)
except: pass
_OUTDOOR_360 = {"bicycle", "flowers", "garden", "stump", "treehill"}
if _explicit_res is not None and _explicit_res > 0:
_res = _explicit_res
elif _scene is not None:
_res = 4 if _scene in _OUTDOOR_360 else 2
else:
_res = None
try:
if _res is not None:
self.dataset.resolution = _res
print("[res-fix] scene=%s explicit=%s -> res=%s (%s)" % (_scene, _explicit_res, _res, __file__))
except Exception as _e:
print("[res-fix] FAILED:", _e)
# INJECTED_RES_FIX end
self.scene = Scene(self.dataset, self.gaussians)
self.gaussians.training_setup(self.opt)
bg_color = [1, 1, 1] if self.dataset.white_background else [0, 0, 0]
self.background = torch.tensor(bg_color, dtype=torch.float32, device="cuda")
self.viewpoint_stack = self.scene.getTrainCameras().copy()
self.last_n_gaussians = len(self.gaussians.get_xyz)
self.KNN_index = None
def train_iteration(self, step):
self.gaussians.update_learning_rate(step)
if step % 1000 == 0:
self.gaussians.oneupSHdegree()
if not self.viewpoint_stack:
self.viewpoint_stack = self.scene.getTrainCameras().copy()
viewpoint_cam = self.viewpoint_stack.pop(random.randint(0, len(self.viewpoint_stack) - 1))
bg = torch.rand((3), device="cuda") if self.opt.random_background else self.background
render_pkg = native_render(viewpoint_cam, self.gaussians, self.pipe, bg)
image = render_pkg["render"]
viewspace_point_tensor = render_pkg["viewspace_points"]
visibility_filter = render_pkg["visibility_filter"]
radii = render_pkg["radii"]
gt_image = viewpoint_cam.original_image.cuda()
Ll1 = l1_loss(image, gt_image)
ssim_value = ssim(image, gt_image)
loss_target = (1.0 - self.opt.lambda_dssim) * Ll1
loss_parasitic_ssim = self.opt.lambda_dssim * (1.0 - ssim_value)
pair_d_loss = torch.tensor(0.0, device="cuda")
pair_normal_loss = torch.tensor(0.0, device="cuda")
if self.KNN_index is not None and visibility_filter.sum() > 0:
mask = torch.logical_and(visibility_filter, self.gaussians.get_type.squeeze() == 1)
if mask.sum() > 0:
pair_d_loss_raw, pair_normal_loss_raw = SurfaceAlign()(
self.gaussians.get_xyz,
self.gaussians.get_xyz_id.contiguous(),
self.gaussians.get_rotation,
self.KNN_index[mask]
)
pair_d_loss = 0.05 * torch.mean(pair_d_loss_raw)
pair_normal_loss = 0.01 * torch.mean(pair_normal_loss_raw)
loss_parasitic = loss_parasitic_ssim + pair_d_loss + pair_normal_loss
loss = loss_target + loss_parasitic
grad_cos_sim = 0.0
parasitic_ratio = 0.0
stats = {}
if self.track_decoupling and step % 100 == 0:
self.gaussians.optimizer.zero_grad(set_to_none=True)
loss_target.backward(retain_graph=True)
grads_target = {}
for group in self.gaussians.optimizer.param_groups:
name = group["name"]
p = group["params"][0]
if p.grad is not None:
grads_target[name] = p.grad.clone()
self.gaussians.optimizer.zero_grad(set_to_none=True)
loss_parasitic.backward(retain_graph=True)
grads_parasitic = {}
for group in self.gaussians.optimizer.param_groups:
name = group["name"]
p = group["params"][0]
if p.grad is not None:
grads_parasitic[name] = p.grad.clone()
gt_xyz = grads_target.get("xyz", torch.zeros(1, device="cuda"))
gp_xyz = grads_parasitic.get("xyz", torch.zeros(1, device="cuda"))
valid_mask = (torch.norm(gt_xyz, dim=1) > 0) & (torch.norm(gp_xyz, dim=1) > 0)
if valid_mask.any():
grad_cos_sim = float(F.cosine_similarity(gt_xyz[valid_mask], gp_xyz[valid_mask], dim=1).mean())
parasitic_ratio = float(torch.norm(gp_xyz, dim=1).mean() / (torch.norm(gt_xyz, dim=1).mean() + 1e-7))
group_mapping = {
"spatial": ["xyz"],
"geometry": ["scaling", "rotation"],
"opacity": ["opacity"],
"appearance": ["f_dc", "f_rest"]
}
for super_group, sub_names in group_mapping.items():
u_t_list = []
u_p_list = []
for name in sub_names:
for group in self.gaussians.optimizer.param_groups:
if group["name"] == name:
p = group["params"][0]
state = self.gaussians.optimizer.state.get(p)
vt = state.get("exp_avg_sq", torch.zeros_like(p)) if state else torch.zeros_like(p)
lr = group["lr"]
gt = grads_target.get(name, torch.zeros_like(p))
gp = grads_parasitic.get(name, torch.zeros_like(p))
step_t = (lr / (torch.sqrt(vt) + 1e-15)) * gt
step_p = (lr / (torch.sqrt(vt) + 1e-15)) * gp
u_t_list.append(step_t.reshape(-1))
u_p_list.append(step_p.reshape(-1))
if u_t_list and u_p_list:
U_T = torch.cat(u_t_list)
U_P = torch.cat(u_p_list)
if U_T.norm() > 0 and U_P.norm() > 0:
cos = float(F.cosine_similarity(U_T.unsqueeze(0), U_P.unsqueeze(0)))
r = float(U_P.norm() / (U_T.norm() + U_P.norm() + 1e-7))
ti = r * max(0.0, -cos)
else:
ti = 0.0
stats[f"sti_{super_group}"] = ti
else:
stats[f"sti_{super_group}"] = 0.0
self.gaussians.optimizer.zero_grad(set_to_none=True)
loss.backward()
else:
loss.backward()
with torch.no_grad():
if step < self.opt.densify_until_iter:
self.gaussians.max_radii2D[visibility_filter] = torch.max(self.gaussians.max_radii2D[visibility_filter], radii[visibility_filter])
self.gaussians.add_densification_stats(viewspace_point_tensor, visibility_filter)
if step > self.opt.densify_from_iter and step % self.opt.densification_interval == 0:
size_threshold = 20 if step > self.opt.opacity_reset_interval else None
self.gaussians.densify_and_prune(self.opt.densify_grad_threshold, 0.05, self.scene.cameras_extent, size_threshold)
self.KNN_index = self.gaussians.findKNN()
if step % self.opt.opacity_reset_interval == 0 or (self.dataset.white_background and step == self.opt.densify_from_iter):
self.gaussians.reset_opacity()
elif step > self.opt.densify_until_iter:
if step % 3000 == 0:
self.KNN_index = self.gaussians.findKNN()
if step <= self.opt.iterations:
self.gaussians.optimizer.step()
self.gaussians.optimizer.zero_grad(set_to_none=True)
num_gaussians = self.gaussians.get_xyz.shape[0]
metrics = {
"loss": float(loss), "loss_l1": float(loss_target), "loss_ssim": float(loss_parasitic_ssim),
"loss_pair_d": float(pair_d_loss), "loss_pair_normal": float(pair_normal_loss),
"num_gaussians": int(num_gaussians), "delta_N": int(num_gaussians - self.last_n_gaussians),
"peak_vram_GB": float(torch.cuda.max_memory_allocated() / (1024 ** 3)),
"grad_cos_sim": float(grad_cos_sim), "parasitic_ratio": float(parasitic_ratio)
}
metrics.update(stats)
self.last_n_gaussians = num_gaussians
histograms = {}
if step % 1000 == 0:
histograms["opacity"] = torch.sigmoid(self.gaussians._opacity).clone().detach()
scales = torch.exp(self.gaussians._scaling).clone().detach()
histograms["scaling"] = scales
scales_2d = scales[:, :2] if scales.shape[1] >= 2 else scales.unsqueeze(-1).expand(-1, 2)
gamma = scales_2d.max(dim=-1)[0] / (scales_2d.min(dim=-1)[0] + 1e-7)
histograms["anisotropy"] = gamma
histograms["sh_dc_mag"] = self.gaussians._features_dc.detach().norm(dim=-1)
return metrics, histograms
def render(self, camera):
with torch.no_grad():
render_pkg = native_render(camera, self.gaussians, self.pipe, self.background)
return {"image": render_pkg["render"], "depth": render_pkg.get("render_depth", None)}
def save(self, save_dir, step):
self.scene.save(step)
def load(self, model_path, iteration):
self.gaussians.load_ply(os.path.join(model_path, 'point_cloud', f'iteration_{iteration}', 'point_cloud.ply'))
def get_spatial_centers(self):
return self.gaussians._xyz
def compute_physical_metrics(self, cameras=None):
metrics = {}
with torch.no_grad():
raw_scales = self.gaussians._scaling
scales = torch.exp(raw_scales)
scales_2d = scales[:, :2] if scales.dim() > 1 and scales.shape[1] >= 2 else scales.unsqueeze(-1).expand(-1, 2)
max_S, _ = torch.max(scales_2d, dim=1)
min_S, _ = torch.min(scales_2d, dim=1)
gamma = max_S / (min_S + 1e-7)
metrics["gamma_median"] = float(torch.median(gamma))
metrics["gamma_90th_percentile"] = float(torch.quantile(gamma, 0.90))
metrics["scale_mean"] = float(torch.mean(scales_2d))
metrics["alpha_mean"] = float(torch.mean(torch.sigmoid(self.gaussians._opacity)))
dc, rest = self.gaussians._features_dc, self.gaussians._features_rest
if rest is not None and rest.shape[1] > 0:
metrics["sh_energy_ratio"] = float(rest.norm(dim=-1).mean() / (dc.norm(dim=-1).mean() + 1e-7))
if cameras is not None and len(cameras) > 0:
view_dirs = []
for c in cameras:
view_dirs.append(c.world_view_transform[:3, 2].tolist())
view_dirs = F.normalize(torch.tensor(view_dirs, dtype=torch.float32, device="cuda"), dim=1)
rots = F.normalize(self.gaussians._rotation.clone(), dim=1)
w, x, y, z = rots.unbind(dim=-1)
normals = F.normalize(torch.stack([2*(x*z + w*y), 2*(y*z - w*x), 1-2*(x*x + y*y)], dim=-1), dim=1)
max_cos, _ = torch.max(torch.abs(torch.matmul(normals, view_dirs.T)), dim=1)
metrics["billboard_bias_ratio"] = float((max_cos > 0.90).float().mean())
return metrics
def evaluate_spatial_field(self, query_points: torch.Tensor, cameras=None) -> torch.Tensor:
with torch.no_grad():
V = query_points.shape[0]
densities = torch.zeros(V, device="cuda")
xyz = self.gaussians._xyz
opacities = torch.sigmoid(self.gaussians._opacity).squeeze()
scales = torch.exp(self.gaussians._scaling)
sigma_sq = (scales[:, :2].max(dim=1)[0].pow(2)) if scales.shape[1] >= 2 else scales.squeeze().pow(2)
N_gaussians = xyz.shape[0]
chunk_size = max(1, 30_000_000 // (N_gaussians + 1))
for i in range(0, V, chunk_size):
end = min(i + chunk_size, V)
dist_sq = torch.cdist(query_points[i:end], xyz, p=2).pow(2)
weights = torch.exp(-0.5 * dist_sq / (sigma_sq.unsqueeze(0) + 1e-7))
densities[i:end] = torch.sum(weights * opacities.unsqueeze(0), dim=1)
return densities
|