reduced-3dgs 1.8.16__tar.gz → 1.8.18__tar.gz

{reduced_3dgs-1.8.16 → reduced_3dgs-1.8.18}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: reduced_3dgs
-Version: 1.8.16
+Version: 1.8.18
 Summary: Refactored code for the paper "Reducing the Memory Footprint of 3D Gaussian Splatting"
 Home-page: https://github.com/yindaheng98/reduced-3dgs
 Author: yindaheng98

reduced_3dgs-1.8.18/reduced_3dgs/importance/__init__.py

@@ -0,0 +1,2 @@
+from .trainer import ImportancePruner, BaseImportancePruningTrainer, ImportancePrunerWrapper
+from .combinations import BaseImportancePrunerInDensifyTrainer, ImportancePruningTrainer, ImportancePrunerInDensifyTrainer

reduced_3dgs-1.8.18/reduced_3dgs/importance/combinations.py

@@ -0,0 +1,42 @@
+from typing import List
+from gaussian_splatting import Camera, GaussianModel
+from gaussian_splatting.dataset import TrainableCameraDataset
+from gaussian_splatting.trainer import DepthTrainerWrapper, NoopDensifier, DensificationTrainerWrapper
+from .trainer import ImportancePruner, BaseImportancePruningTrainer
+
+
+def BaseImportancePrunerInDensifyTrainer(
+        model: GaussianModel,
+        scene_extent: float,
+        dataset: List[Camera],
+        importance_prune_from_iter=1000,
+        importance_prune_until_iter=15000,
+        importance_prune_interval=100,
+        *args, **kwargs):
+    return DensificationTrainerWrapper(
+        lambda model, scene_extent: ImportancePruner(
+            NoopDensifier(model),
+            dataset,
+            importance_prune_from_iter=importance_prune_from_iter,
+            importance_prune_until_iter=importance_prune_until_iter,
+            importance_prune_interval=importance_prune_interval,
+        ),
+        model,
+        scene_extent,
+        *args, **kwargs
+    )
+
+
+# Depth trainer
+
+
+def DepthImportancePruningTrainer(model: GaussianModel, scene_extent: float, dataset: TrainableCameraDataset, *args, **kwargs):
+    return DepthTrainerWrapper(BaseImportancePruningTrainer, model, scene_extent, *args, dataset=dataset, **kwargs)
+
+
+def DepthImportancePrunerInDensifyTrainer(model: GaussianModel, scene_extent: float, dataset: TrainableCameraDataset, *args, **kwargs):
+    return DepthTrainerWrapper(BaseImportancePrunerInDensifyTrainer, model, scene_extent, *args, dataset=dataset, **kwargs)
+
+
+ImportancePruningTrainer = DepthImportancePruningTrainer
+ImportancePrunerInDensifyTrainer = DepthImportancePrunerInDensifyTrainer

reduced_3dgs-1.8.18/reduced_3dgs/importance/trainer.py

@@ -0,0 +1,147 @@
+import math
+from typing import Callable, List
+import torch
+
+from gaussian_splatting import Camera, GaussianModel
+from gaussian_splatting.trainer import AbstractDensifier, DensifierWrapper, DensificationTrainer, NoopDensifier
+from gaussian_splatting.dataset import CameraDataset
+from .diff_gaussian_rasterization import GaussianRasterizationSettings, GaussianRasterizer
+
+
+def count_render(self: GaussianModel, viewpoint_camera: Camera):
+    """
+    Render the scene.
+
+    Background tensor (bg_color) must be on GPU!
+    """
+    # Create zero tensor. We will use it to make pytorch return gradients of the 2D (screen-space) means
+    screenspace_points = torch.zeros_like(self.get_xyz, dtype=self.get_xyz.dtype, requires_grad=True, device=self._xyz.device) + 0
+    try:
+        screenspace_points.retain_grad()
+    except:
+        pass
+
+    # Set up rasterization configuration
+    tanfovx = math.tan(viewpoint_camera.FoVx * 0.5)
+    tanfovy = math.tan(viewpoint_camera.FoVy * 0.5)
+
+    raster_settings = GaussianRasterizationSettings(
+        image_height=int(viewpoint_camera.image_height),
+        image_width=int(viewpoint_camera.image_width),
+        tanfovx=tanfovx,
+        tanfovy=tanfovy,
+        bg=viewpoint_camera.bg_color.to(self._xyz.device),
+        scale_modifier=self.scale_modifier,
+        viewmatrix=viewpoint_camera.world_view_transform,
+        projmatrix=viewpoint_camera.full_proj_transform,
+        sh_degree=self.active_sh_degree,
+        campos=viewpoint_camera.camera_center,
+        prefiltered=False,
+        debug=self.debug,
+        f_count=True,
+    )
+
+    rasterizer = GaussianRasterizer(raster_settings=raster_settings)
+    means3D = self.get_xyz
+    means2D = screenspace_points
+    opacity = self.get_opacity
+
+    scales = self.get_scaling
+    rotations = self.get_rotation
+
+    shs = self.get_features
+
+    # Rasterize visible Gaussians to image, obtain their radii (on screen).
+    gaussians_count, opacity_important_score, T_alpha_important_score, rendered_image, radii = rasterizer(
+        means3D=means3D,
+        means2D=means2D,
+        shs=shs,
+        colors_precomp=None,
+        opacities=opacity,
+        scales=scales,
+        rotations=rotations,
+        cov3D_precomp=None)
+
+    # Those Gaussians that were frustum culled or had a radius of 0 were not visible.
+    # They will be excluded from value updates used in the splitting criteria.
+    return {
+        "render": rendered_image,
+        "viewspace_points": screenspace_points,
+        "visibility_filter": radii > 0,
+        "radii": radii,
+        "gaussians_count": gaussians_count,
+        "opacity_important_score": opacity_important_score,
+        "T_alpha_important_score": T_alpha_important_score
+    }
+
+
+def prune_gaussians(model: GaussianModel, dataset: CameraDataset):
+    gaussian_count = torch.zeros(model.get_xyz.shape[0], device=model.get_xyz.device, dtype=torch.int)
+    opacity_important_score = torch.zeros(model.get_xyz.shape[0], device=model.get_xyz.device, dtype=torch.float)
+    T_alpha_important_score = torch.zeros(model.get_xyz.shape[0], device=model.get_xyz.device, dtype=torch.float)
+    for camera in dataset:
+        out = count_render(model, camera)
+        gaussian_count += out["gaussians_count"]
+        opacity_important_score += out["opacity_important_score"]
+        T_alpha_important_score += out["T_alpha_important_score"]
+    return None
+
+
+class ImportancePruner(DensifierWrapper):
+    def __init__(
+            self, base_densifier: AbstractDensifier,
+            dataset: CameraDataset,
+            importance_prune_from_iter=1000,
+            importance_prune_until_iter=15000,
+            importance_prune_interval: int = 100,
+    ):
+        super().__init__(base_densifier)
+        self.dataset = dataset
+        self.importance_prune_from_iter = importance_prune_from_iter
+        self.importance_prune_until_iter = importance_prune_until_iter
+        self.importance_prune_interval = importance_prune_interval
+
+    def densify_and_prune(self, loss, out, camera, step: int):
+        ret = super().densify_and_prune(loss, out, camera, step)
+        if self.importance_prune_from_iter <= step <= self.importance_prune_until_iter and step % self.importance_prune_interval == 0:
+            remove_mask = prune_gaussians(self.model, self.dataset)
+            ret = ret._replace(remove_mask=remove_mask if ret.remove_mask is None else torch.logical_or(ret.remove_mask, remove_mask))
+        return ret
+
+
+def ImportancePrunerWrapper(
+        base_densifier_constructor: Callable[..., AbstractDensifier],
+        model: GaussianModel,
+        scene_extent: float,
+        dataset: List[Camera],
+        importance_prune_from_iter=1000,
+        importance_prune_until_iter=15000,
+        importance_prune_interval: int = 100,
+        *args, **kwargs):
+    return ImportancePruner(
+        base_densifier_constructor(model, scene_extent, *args, **kwargs),
+        dataset,
+        importance_prune_from_iter=importance_prune_from_iter,
+        importance_prune_until_iter=importance_prune_until_iter,
+        importance_prune_interval=importance_prune_interval,
+    )
+
+
+def BaseImportancePruningTrainer(
+        model: GaussianModel,
+        scene_extent: float,
+        dataset: List[Camera],
+        importance_prune_from_iter=1000,
+        importance_prune_until_iter=15000,
+        importance_prune_interval: int = 100,
+        *args, **kwargs):
+    return DensificationTrainer(
+        model, scene_extent,
+        ImportancePruner(
+            NoopDensifier(model),
+            dataset,
+            importance_prune_from_iter=importance_prune_from_iter,
+            importance_prune_until_iter=importance_prune_until_iter,
+            importance_prune_interval=importance_prune_interval,
+        ), *args, **kwargs
+    )

{reduced_3dgs-1.8.16 → reduced_3dgs-1.8.18}/reduced_3dgs/pruning/combinations.py

@@ -21,7 +21,7 @@ def BasePrunerInDensifyTrainer(
     return DensificationTrainerWrapper(
         lambda model, scene_extent: BasePruner(
             NoopDensifier(model),
-            model, dataset,
+            dataset,
             prune_from_iter=prune_from_iter,
             prune_until_iter=prune_until_iter,
             prune_interval=prune_interval,

{reduced_3dgs-1.8.16 → reduced_3dgs-1.8.18}/reduced_3dgs/pruning/trainer.py

@@ -82,7 +82,7 @@ def mercy_gaussians(
 class BasePruner(DensifierWrapper):
     def __init__(
             self, base_densifier: AbstractDensifier,
-            model: GaussianModel, dataset: List[Camera],
+            dataset: List[Camera],
             prune_from_iter=1000,
             prune_until_iter=15000,
             prune_interval: int = 100,
@@ -91,7 +91,6 @@ class BasePruner(DensifierWrapper):
             mercy_minimum=3,
             mercy_type='redundancy_opacity'):
         super().__init__(base_densifier)
-        self._model = model
         self.dataset = dataset
         self.prune_from_iter = prune_from_iter
         self.prune_until_iter = prune_until_iter
@@ -101,10 +100,6 @@ class BasePruner(DensifierWrapper):
         self.mercy_minimum = mercy_minimum
         self.mercy_type = mercy_type
 
-    @property
-    def model(self) -> GaussianModel:
-        return self._model
-
     def densify_and_prune(self, loss, out, camera, step: int):
         ret = super().densify_and_prune(loss, out, camera, step)
         if self.prune_from_iter <= step <= self.prune_until_iter and step % self.prune_interval == 0:
@@ -128,8 +123,8 @@ def BasePruningTrainer(
     return DensificationTrainer(
         model, scene_extent,
         BasePruner(
-            NoopDensifier(),
-            model, dataset,
+            NoopDensifier(model),
+            dataset,
             prune_from_iter=prune_from_iter,
             prune_until_iter=prune_until_iter,
             prune_interval=prune_interval,

{reduced_3dgs-1.8.16 → reduced_3dgs-1.8.18}/reduced_3dgs.egg-info/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: reduced_3dgs
-Version: 1.8.16
+Version: 1.8.18
 Summary: Refactored code for the paper "Reducing the Memory Footprint of 3D Gaussian Splatting"
 Home-page: https://github.com/yindaheng98/reduced-3dgs
 Author: yindaheng98

{reduced_3dgs-1.8.16 → reduced_3dgs-1.8.18}/reduced_3dgs.egg-info/SOURCES.txt

@@ -10,11 +10,12 @@ reduced_3dgs.egg-info/SOURCES.txt
 reduced_3dgs.egg-info/dependency_links.txt
 reduced_3dgs.egg-info/requires.txt
 reduced_3dgs.egg-info/top_level.txt
+reduced_3dgs/importance/__init__.py
+reduced_3dgs/importance/combinations.py
+reduced_3dgs/importance/trainer.py
 reduced_3dgs/pruning/__init__.py
 reduced_3dgs/pruning/combinations.py
 reduced_3dgs/pruning/trainer.py
-reduced_3dgs/pruning/importance/__init__.py
-reduced_3dgs/pruning/importance/trainer.py
 reduced_3dgs/quantization/__init__.py
 reduced_3dgs/quantization/abc.py
 reduced_3dgs/quantization/exclude_zeros.py

{reduced_3dgs-1.8.16 → reduced_3dgs-1.8.18}/setup.py

@@ -48,7 +48,7 @@ importance_sources = [
     "rasterize_points.cu",
     "ext.cpp"]
 importance_packages = {
-    'reduced_3dgs.pruning.importance.diff_gaussian_rasterization': 'submodules/gaussian-importance/diff_gaussian_rasterization',
+    'reduced_3dgs.importance.diff_gaussian_rasterization': 'submodules/gaussian-importance/diff_gaussian_rasterization',
 }
 
 cxx_compiler_flags = []
@@ -60,7 +60,7 @@ if os.name == 'nt':
 
 setup(
     name="reduced_3dgs",
-    version='1.8.16',
+    version='1.8.18',
     author='yindaheng98',
     author_email='yindaheng98@gmail.com',
     url='https://github.com/yindaheng98/reduced-3dgs',
@@ -83,7 +83,7 @@ setup(
             extra_compile_args={"nvcc": nvcc_compiler_flags + ["-I" + os.path.join(os.path.abspath(rasterizor_root), "third_party/glm/")]}
         ),
         CUDAExtension(
-            name="reduced_3dgs.pruning.importance.diff_gaussian_rasterization._C",
+            name="reduced_3dgs.importance.diff_gaussian_rasterization._C",
             sources=[os.path.join(importance_root, source) for source in importance_sources],
             extra_compile_args={"nvcc": nvcc_compiler_flags + ["-I" + os.path.join(os.path.abspath(importance_root), "third_party/glm/")]}
         ),

reduced_3dgs-1.8.16/reduced_3dgs/pruning/importance/__init__.py

@@ -1 +0,0 @@
-from .trainer import ImportancePruningTrainerWrapper, BaseImportancePruningTrainer, ImportancePruningTrainer

reduced_3dgs-1.8.16/reduced_3dgs/pruning/importance/trainer.py

@@ -1,141 +0,0 @@
-import math
-import torch
-
-from gaussian_splatting import Camera, GaussianModel
-from gaussian_splatting.trainer import AbstractTrainer, TrainerWrapper, BaseTrainer, Trainer
-from gaussian_splatting.dataset import CameraDataset
-from .diff_gaussian_rasterization import GaussianRasterizationSettings, GaussianRasterizer
-
-
-def count_render(self: GaussianModel, viewpoint_camera: Camera):
-    """
-    Render the scene.
-
-    Background tensor (bg_color) must be on GPU!
-    """
-    # Create zero tensor. We will use it to make pytorch return gradients of the 2D (screen-space) means
-    screenspace_points = torch.zeros_like(self.get_xyz, dtype=self.get_xyz.dtype, requires_grad=True, device=self._xyz.device) + 0
-    try:
-        screenspace_points.retain_grad()
-    except:
-        pass
-
-    # Set up rasterization configuration
-    tanfovx = math.tan(viewpoint_camera.FoVx * 0.5)
-    tanfovy = math.tan(viewpoint_camera.FoVy * 0.5)
-
-    raster_settings = GaussianRasterizationSettings(
-        image_height=int(viewpoint_camera.image_height),
-        image_width=int(viewpoint_camera.image_width),
-        tanfovx=tanfovx,
-        tanfovy=tanfovy,
-        bg=viewpoint_camera.bg_color.to(self._xyz.device),
-        scale_modifier=self.scale_modifier,
-        viewmatrix=viewpoint_camera.world_view_transform,
-        projmatrix=viewpoint_camera.full_proj_transform,
-        sh_degree=self.active_sh_degree,
-        campos=viewpoint_camera.camera_center,
-        prefiltered=False,
-        debug=self.debug,
-        f_count=True,
-    )
-
-    rasterizer = GaussianRasterizer(raster_settings=raster_settings)
-    means3D = self.get_xyz
-    means2D = screenspace_points
-    opacity = self.get_opacity
-
-    scales = self.get_scaling
-    rotations = self.get_rotation
-
-    shs = self.get_features
-
-    # Rasterize visible Gaussians to image, obtain their radii (on screen).
-    gaussians_count, opacity_important_score, T_alpha_important_score, rendered_image, radii = rasterizer(
-        means3D=means3D,
-        means2D=means2D,
-        shs=shs,
-        colors_precomp=None,
-        opacities=opacity,
-        scales=scales,
-        rotations=rotations,
-        cov3D_precomp=None)
-
-    # Those Gaussians that were frustum culled or had a radius of 0 were not visible.
-    # They will be excluded from value updates used in the splitting criteria.
-    return {
-        "render": rendered_image,
-        "viewspace_points": screenspace_points,
-        "visibility_filter": radii > 0,
-        "radii": radii,
-        "gaussians_count": gaussians_count,
-        "opacity_important_score": opacity_important_score,
-        "T_alpha_important_score": T_alpha_important_score
-    }
-
-
-class ImportancePruner(TrainerWrapper):
-    def __init__(
-            self, base_trainer: AbstractTrainer,
-            dataset: CameraDataset,
-            importance_prune_at_steps=[15000],
-    ):
-        super().__init__(base_trainer)
-        self.dataset = dataset
-        self.importance_prune_at_steps = importance_prune_at_steps
-
-    def optim_step(self):
-        ret = super().optim_step()
-        if self.curr_step in self.importance_prune_at_steps:
-            gaussian_count = torch.zeros(self.model.get_xyz.shape[0], device=self.model.get_xyz.device, dtype=torch.int)
-            opacity_important_score = torch.zeros(self.model.get_xyz.shape[0], device=self.model.get_xyz.device, dtype=torch.float)
-            T_alpha_important_score = torch.zeros(self.model.get_xyz.shape[0], device=self.model.get_xyz.device, dtype=torch.float)
-            for camera in self.dataset:
-                out = count_render(self.model, camera)
-                gaussian_count += out["gaussians_count"]
-                opacity_important_score += out["opacity_important_score"]
-                T_alpha_important_score += out["T_alpha_important_score"]
-            pass
-        return ret
-
-
-def ImportancePruningTrainerWrapper(
-    base_trainer_constructor,
-        model: GaussianModel,
-        scene_extent: float,
-        dataset: CameraDataset,
-        importance_prune_at_steps=[15000],
-        *args, **kwargs):
-    return ImportancePruner(
-        base_trainer_constructor(model, scene_extent, dataset, *args, **kwargs),
-        dataset,
-        importance_prune_at_steps=importance_prune_at_steps,
-    )
-
-
-def BaseImportancePruningTrainer(
-    model: GaussianModel,
-        scene_extent: float,
-        dataset: CameraDataset,
-        importance_prune_at_steps=[15000],
-        *args, **kwargs):
-    return ImportancePruningTrainerWrapper(
-        lambda model, scene_extent, dataset, *args, **kwargs: BaseTrainer(model, scene_extent, *args, **kwargs),
-        model, scene_extent, dataset,
-        importance_prune_at_steps=importance_prune_at_steps,
-        *args, **kwargs,
-    )
-
-
-def ImportancePruningTrainer(
-    model: GaussianModel,
-        scene_extent: float,
-        dataset: CameraDataset,
-        importance_prune_at_steps=[15000],
-        *args, **kwargs):
-    return ImportancePruningTrainerWrapper(
-        lambda model, scene_extent, dataset, *args, **kwargs: Trainer(model, scene_extent, *args, **kwargs),
-        model, scene_extent, dataset,
-        importance_prune_at_steps=importance_prune_at_steps,
-        *args, **kwargs,
-    )