images for debugging TexturesUV

Summary: New methods to directly plot a TexturesUV map with its used points, using PIL and matplotlib.

Reviewed By: gkioxari

Differential Revision: D23782968

fbshipit-source-id: 692970857b5be13a35a3175dc82ac03963a73555

docs/tutorials/render_textured_meshes.ipynb

@@ -93,7 +93,7 @@
     "\n",
     "# Data structures and functions for rendering\n",
     "from pytorch3d.structures import Meshes\n",
-    "from pytorch3d.vis import AxisArgs, plot_batch_individually, plot_scene\n",
+    "from pytorch3d.vis import AxisArgs, plot_batch_individually, plot_scene, texturesuv_image_matplotlib\n",
     "from pytorch3d.renderer import (\n",
     "    look_at_view_transform,\n",
     "    FoVPerspectiveCameras, \n",
@@ -236,8 +236,7 @@
     "obj_filename = os.path.join(DATA_DIR, \"cow_mesh/cow.obj\")\n",
     "\n",
     "# Load obj file\n",
-    "mesh = load_objs_as_meshes([obj_filename], device=device)\n",
-    "texture_image=mesh.textures.maps_padded()"
+    "mesh = load_objs_as_meshes([obj_filename], device=device)"
    ]
   },
   {
@@ -265,9 +264,29 @@
    "outputs": [],
    "source": [
     "plt.figure(figsize=(7,7))\n",
+    "texture_image=mesh.textures.maps_padded()\n",
     "plt.imshow(texture_image.squeeze().cpu().numpy())\n",
     "plt.grid(\"off\");\n",
-    "plt.axis('off');"
+    "plt.axis(\"off\");"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "PyTorch3D has a built-in way to view the texture map with matplotlib along with the points on the map corresponding to vertices. There is also a method, texturesuv_image_PIL, to get a similar image which can be saved to a file."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "plt.figure(figsize=(7,7))\n",
+    "texturesuv_image_matplotlib(mesh.textures, subsample=None)\n",
+    "plt.grid(\"off\");\n",
+    "plt.axis(\"off\");"
    ]
   },
   {

pytorch3d/renderer/mesh/textures.py

@@ -1174,6 +1174,42 @@ class TexturesUV(TexturesBase):
             padding_mode=self.padding_mode,
         )
 
+    def centers_for_image(self, index):
+        """
+        Return the locations in the texture map which correspond to the given
+        verts_uvs, for one of the meshes. This is potentially useful for
+        visualizing the data. See the texturesuv_image_matplotlib and
+        texturesuv_image_PIL functions.
+
+        Args:
+            index: batch index of the mesh whose centers to return.
+
+        Returns:
+            centers: coordinates of points in the texture image
+                - a FloatTensor of shape (V,2)
+        """
+        if self._N != 1:
+            raise ValueError(
+                "This function only supports plotting textures for one mesh."
+            )
+        texture_image = self.maps_padded()
+        verts_uvs = self.verts_uvs_list()[index][None]
+        _, H, W, _3 = texture_image.shape
+        coord1 = torch.arange(W).expand(H, W)
+        coord2 = torch.arange(H)[:, None].expand(H, W)
+        coords = torch.stack([coord1, coord2])[None]
+        with torch.no_grad():
+            # Get xy cartesian coordinates based on the uv coordinates
+            centers = F.grid_sample(
+                torch.flip(coords.to(texture_image), [2]),
+                # Convert from [0, 1] -> [-1, 1] range expected by grid sample
+                verts_uvs[:, None] * 2.0 - 1,
+                align_corners=self.align_corners,
+                padding_mode=self.padding_mode,
+            ).cpu()
+            centers = centers[0, :, 0].T
+        return centers
+
 
 class TexturesVertex(TexturesBase):
     def __init__(

pytorch3d/vis/__init__.py

@@ -1,6 +1,7 @@
 # Copyright (c) Facebook, Inc. and its affiliates. All rights reserved.
 
 from .plotly_vis import AxisArgs, Lighting, plot_batch_individually, plot_scene
+from .texture_vis import texturesuv_image_matplotlib, texturesuv_image_PIL
 
 
 __all__ = [k for k in globals().keys() if not k.startswith("_")]

pytorch3d/vis/texture_vis.py

@@ -0,0 +1,104 @@
+# Copyright (c) Facebook, Inc. and its affiliates. All rights reserved.
+from typing import Optional
+
+import numpy as np
+from PIL import Image, ImageDraw
+from pytorch3d.renderer.mesh import TexturesUV
+
+
+def texturesuv_image_matplotlib(
+    texture: TexturesUV,
+    *,
+    texture_index: int = 0,
+    radius: float = 1,
+    color=(1.0, 0.0, 0.0),
+    subsample: Optional[int] = 10000,
+    origin: str = "upper",
+):
+    """
+    Plot the texture image for one element of a TexturesUV with
+    matplotlib together with verts_uvs positions circled.
+    In particular a value in verts_uvs which is never referenced
+    in faces_uvs will still be plotted.
+    This is for debugging purposes, e.g. to align the map with
+    the uv coordinates. In particular, matplotlib
+    is used which is not an official dependency of PyTorch3D.
+
+    Args:
+        texture: a TexturesUV object with one mesh
+        texture_index: index in the batch to plot
+        radius: plotted circle radius in pixels
+        color: any matplotlib-understood color for the circles.
+        subsample: if not None, number of points to plot.
+                Otherwise all points are plotted.
+        origin: "upper" or "lower" like matplotlib.imshow
+    """
+
+    import matplotlib.pyplot as plt
+    from matplotlib.patches import Circle
+
+    texture_image = texture.maps_padded()
+    centers = texture.centers_for_image(index=texture_index).numpy()
+
+    ax = plt.gca()
+    ax.imshow(texture_image[texture_index].detach().cpu().numpy(), origin=origin)
+
+    n_points = centers.shape[0]
+    if subsample is None or n_points <= subsample:
+        indices = range(n_points)
+    else:
+        indices = np.random.choice(n_points, subsample, replace=False)
+    for i in indices:
+        # setting clip_on=False makes it obvious when
+        # we have UV coordinates outside the correct range
+        ax.add_patch(Circle(centers[i], radius, color=color, clip_on=False))
+
+
+def texturesuv_image_PIL(
+    texture: TexturesUV,
+    *,
+    texture_index: int = 0,
+    radius: float = 1,
+    color="red",
+    subsample: Optional[int] = 10000,
+):
+    """
+    Return a PIL image of the texture image of one element of the batch
+    from a TexturesUV, together with the verts_uvs positions circled.
+    In particular a value in verts_uvs which is never referenced
+    in faces_uvs will still be plotted.
+    This is for debugging purposes, e.g. to align the map with
+    the uv coordinates. In particular, matplotlib
+    is used which is not an official dependency of PyTorch3D.
+
+    Args:
+        texture: a TexturesUV object with one mesh
+        texture_index: index in the batch to plot
+        radius: plotted circle radius in pixels
+        color: any PIL-understood color for the circles.
+        subsample: if not None, number of points to plot.
+                Otherwise all points are plotted.
+
+    Returns:
+        PIL Image object.
+    """
+
+    centers = texture.centers_for_image(index=texture_index).numpy()
+    texture_image = texture.maps_padded()
+    texture_array = (texture_image[texture_index] * 255).cpu().numpy().astype(np.uint8)
+
+    image = Image.fromarray(texture_array)
+    draw = ImageDraw.Draw(image)
+
+    n_points = centers.shape[0]
+    if subsample is None or n_points <= subsample:
+        indices = range(n_points)
+    else:
+        indices = np.random.choice(n_points, subsample, replace=False)
+
+    for i in indices:
+        x = centers[i][0]
+        y = centers[i][1]
+        draw.ellipse([(x - radius, y - radius), (x + radius, y + radius)], fill=color)
+
+    return image

tests/test_texturing.py

@@ -2,10 +2,13 @@
 
 
 import unittest
+from pathlib import Path
 
+import numpy as np
 import torch
 import torch.nn.functional as F
 from common_testing import TestCaseMixin
+from PIL import Image
 from pytorch3d.renderer.mesh.rasterizer import Fragments
 from pytorch3d.renderer.mesh.textures import (
     TexturesAtlas,
@@ -15,9 +18,14 @@ from pytorch3d.renderer.mesh.textures import (
     pack_rectangles,
 )
 from pytorch3d.structures import Meshes, list_to_packed, packed_to_list
+from pytorch3d.vis import texturesuv_image_PIL
 from test_meshes import TestMeshes
 
 
+DEBUG = False
+DATA_DIR = Path(__file__).resolve().parent / "data"
+
+
 def tryindex(self, index, tex, meshes, source):
     tex2 = tex[index]
     meshes2 = meshes[index]
@@ -471,6 +479,10 @@ class TestTexturesAtlas(TestCaseMixin, unittest.TestCase):
 
 
 class TestTexturesUV(TestCaseMixin, unittest.TestCase):
+    def setUp(self) -> None:
+        super().setUp()
+        torch.manual_seed(42)
+
     def test_sample_textures_uv(self):
         barycentric_coords = torch.tensor(
             [[0.5, 0.3, 0.2], [0.3, 0.6, 0.1]], dtype=torch.float32
@@ -821,6 +833,22 @@ class TestTexturesUV(TestCaseMixin, unittest.TestCase):
         tryindex(self, index, tex, meshes, source)
         tryindex(self, [2, 4], tex, meshes, source)
 
+    def test_png_debug(self):
+        maps = torch.rand(size=(1, 256, 128, 3)) * torch.tensor([0.8, 1, 0.8])
+        verts_uvs = torch.rand(size=(1, 20, 2))
+        faces_uvs = torch.zeros(size=(1, 0, 3), dtype=torch.int64)
+        tex = TexturesUV(maps=maps, faces_uvs=faces_uvs, verts_uvs=verts_uvs)
+
+        image = texturesuv_image_PIL(tex, radius=3)
+        image_out = np.array(image)
+        if DEBUG:
+            image.save(DATA_DIR / "texturesuv_debug_.png")
+
+        with Image.open(DATA_DIR / "texturesuv_debug.png") as image_ref_file:
+            image_ref = np.array(image_ref_file)
+
+        self.assertClose(image_out, image_ref)
+
 
 class TestRectanglePacking(TestCaseMixin, unittest.TestCase):
     def setUp(self) -> None: