update padded in meshes

gkioxari · facebook-github-bot · commit 1fb97f9c8459 · 2020-05-22T22:38:29.000-07:00
Summary:
Three changes to Meshes

1. `num_verts_per_mesh` and `num_faces_per_mesh` are assigned at construction time and are returned without the need for `compute_packed`
2. `update_padded` updates `verts_padded` and shallow copies faces list and faces_padded and existing attributes from construction.
3. `padded_to_packed_idx` does not need `compute_packed`

Reviewed By: nikhilaravi

Differential Revision: D21653674

fbshipit-source-id: dc6815a2e2a925fe4a834fe357919da2b2c14527
diff --git a/pytorch3d/structures/meshes.py b/pytorch3d/structures/meshes.py
@@ -462,9 +462,9 @@ def verts_list(self):
             assert (
                 self._verts_padded is not None
             ), "verts_padded is required to compute verts_list."
-            self._verts_list = [
-                v[0] for v in self._verts_padded.split([1] * self._N, 0)
-            ]
+            self._verts_list = struct_utils.padded_to_list(
+                self._verts_padded, self.num_verts_per_mesh().tolist()
+            )
         return self._verts_list
 
     def faces_list(self):
@@ -478,10 +478,9 @@ def faces_list(self):
             assert (
                 self._faces_padded is not None
             ), "faces_padded is required to compute faces_list."
-            self._faces_list = []
-            for i in range(self._N):
-                valid = self._faces_padded[i].gt(-1).all(1)
-                self._faces_list.append(self._faces_padded[i, valid, :])
+            self._faces_list = struct_utils.padded_to_list(
+                self._faces_padded, self.num_faces_per_mesh().tolist()
+            )
         return self._faces_list
 
     def verts_packed(self):
@@ -525,7 +524,6 @@ def num_verts_per_mesh(self):
         Returns:
             1D tensor of sizes.
         """
-        self._compute_packed()
         return self._num_verts_per_mesh
 
     def faces_packed(self):
@@ -590,7 +588,6 @@ def num_faces_per_mesh(self):
         Returns:
             1D tensor of sizes.
         """
-        self._compute_packed()
         return self._num_faces_per_mesh
 
     def edges_packed(self):
@@ -664,14 +661,13 @@ def verts_padded_to_packed_idx(self):
         Returns:
             1D tensor of indices.
         """
-        self._compute_packed()
         if self._verts_padded_to_packed_idx is not None:
             return self._verts_padded_to_packed_idx
 
         self._verts_padded_to_packed_idx = torch.cat(
             [
                 torch.arange(v, dtype=torch.int64, device=self.device) + i * self._V
-                for (i, v) in enumerate(self._num_verts_per_mesh)
+                for (i, v) in enumerate(self.num_verts_per_mesh())
             ],
             dim=0,
         )
@@ -862,8 +858,8 @@ def _compute_padded(self, refresh: bool = False):
         ):
             return
 
-        verts_list = self._verts_list
-        faces_list = self._faces_list
+        verts_list = self.verts_list()
+        faces_list = self.faces_list()
         assert (
             faces_list is not None and verts_list is not None
         ), "faces_list and verts_list arguments are required"
@@ -943,13 +939,15 @@ def _compute_packed(self, refresh: bool = False):
 
         verts_list_to_packed = struct_utils.list_to_packed(verts_list)
         self._verts_packed = verts_list_to_packed[0]
-        self._num_verts_per_mesh = verts_list_to_packed[1]
+        if not torch.allclose(self.num_verts_per_mesh(), verts_list_to_packed[1]):
+            raise ValueError("The number of verts per mesh should be consistent.")
         self._mesh_to_verts_packed_first_idx = verts_list_to_packed[2]
         self._verts_packed_to_mesh_idx = verts_list_to_packed[3]
 
         faces_list_to_packed = struct_utils.list_to_packed(faces_list)
         faces_packed = faces_list_to_packed[0]
-        self._num_faces_per_mesh = faces_list_to_packed[1]
+        if not torch.allclose(self.num_faces_per_mesh(), faces_list_to_packed[1]):
+            raise ValueError("The number of faces per mesh should be consistent.")
         self._mesh_to_faces_packed_first_idx = faces_list_to_packed[2]
         self._faces_packed_to_mesh_idx = faces_list_to_packed[3]
 
@@ -1328,6 +1326,100 @@ def scale_verts(self, scale):
         new_mesh = self.clone()
         return new_mesh.scale_verts_(scale)
 
+    def update_padded(self, new_verts_padded):
+        """
+        This function allows for an pdate of verts_padded without having to
+        explicitly convert it to the list representation for heterogeneous batches.
+        Returns a Meshes structure with updated padded tensors and copies of the
+        auxiliary tensors at construction time.
+        It updates self._verts_padded with new_verts_padded, and does a
+        shallow copy of (faces_padded, faces_list, num_verts_per_mesh, num_faces_per_mesh).
+        If packed representations are computed in self, they are updated as well.
+
+        Args:
+            new_points_padded: FloatTensor of shape (N, V, 3)
+
+        Returns:
+            Meshes with updated padded representations
+        """
+
+        def check_shapes(x, size):
+            if x.shape[0] != size[0]:
+                raise ValueError("new values must have the same batch dimension.")
+            if x.shape[1] != size[1]:
+                raise ValueError("new values must have the same number of points.")
+            if x.shape[2] != size[2]:
+                raise ValueError("new values must have the same dimension.")
+
+        check_shapes(new_verts_padded, [self._N, self._V, 3])
+
+        new = self.__class__(verts=new_verts_padded, faces=self.faces_padded())
+
+        if new._N != self._N or new._V != self._V or new._F != self._F:
+            raise ValueError("Inconsistent sizes after construction.")
+
+        # overwrite the equisized flag
+        new.equisized = self.equisized
+
+        # overwrite textures if any
+        new.textures = self.textures
+
+        # copy auxiliary tensors
+        copy_tensors = ["_num_verts_per_mesh", "_num_faces_per_mesh", "valid"]
+
+        for k in copy_tensors:
+            v = getattr(self, k)
+            if torch.is_tensor(v):
+                setattr(new, k, v)  # shallow copy
+
+        # shallow copy of faces_list if any, st new.faces_list()
+        # does not re-compute from _faces_padded
+        new._faces_list = self._faces_list
+
+        # update verts/faces packed if they are computed in self
+        if self._verts_packed is not None:
+            copy_tensors = [
+                "_faces_packed",
+                "_verts_packed_to_mesh_idx",
+                "_faces_packed_to_mesh_idx",
+                "_mesh_to_verts_packed_first_idx",
+                "_mesh_to_faces_packed_first_idx",
+            ]
+            for k in copy_tensors:
+                v = getattr(self, k)
+                assert torch.is_tensor(v)
+                setattr(new, k, v)  # shallow copy
+            # update verts_packed
+            pad_to_packed = self.verts_padded_to_packed_idx()
+            new_verts_packed = new_verts_padded.reshape(-1, 3)[pad_to_packed, :]
+            new._verts_packed = new_verts_packed
+            new._verts_padded_to_packed_idx = pad_to_packed
+
+        # update edges packed if they are computed in self
+        if self._edges_packed is not None:
+            copy_tensors = [
+                "_edges_packed",
+                "_edges_packed_to_mesh_idx",
+                "_mesh_to_edges_packed_first_idx",
+                "_faces_packed_to_edges_packed",
+                "_num_edges_per_mesh",
+            ]
+            for k in copy_tensors:
+                v = getattr(self, k)
+                assert torch.is_tensor(v)
+                setattr(new, k, v)  # shallow copy
+
+        # update laplacian if it is compute in self
+        if self._laplacian_packed is not None:
+            new._laplacian_packed = self._laplacian_packed
+
+        assert new._verts_list is None
+        assert new._verts_normals_packed is None
+        assert new._faces_normals_packed is None
+        assert new._faces_areas_packed is None
+
+        return new
+
     # TODO(nikhilar) Move function to utils file.
     def get_bounding_boxes(self):
         """
diff --git a/tests/test_meshes.py b/tests/test_meshes.py
@@ -342,16 +342,11 @@ def test_clone(self):
 
             # Modify tensors in both meshes.
             new_mesh._verts_list[0] = new_mesh._verts_list[0] * 5
-            mesh._num_verts_per_mesh = torch.randint_like(
-                mesh.num_verts_per_mesh(), high=10
-            )
+
             # Check cloned and original Meshes objects do not share tensors.
             self.assertFalse(
                 torch.allclose(new_mesh._verts_list[0], mesh._verts_list[0])
             )
-            self.assertFalse(
-                torch.allclose(mesh.num_verts_per_mesh(), new_mesh.num_verts_per_mesh())
-            )
             self.assertSeparate(new_mesh.verts_packed(), mesh.verts_packed())
             self.assertSeparate(new_mesh.verts_padded(), mesh.verts_padded())
             self.assertSeparate(new_mesh.faces_packed(), mesh.faces_packed())
@@ -690,6 +685,99 @@ def test_split_mesh(self):
         with self.assertRaises(ValueError):
             mesh.split(split_sizes)
 
+    def test_update_padded(self):
+        # Define the test mesh object either as a list or tensor of faces/verts.
+        N = 10
+        for lists_to_tensors in (False, True):
+            for force in (True, False):
+                mesh = TestMeshes.init_mesh(
+                    N, 100, 300, lists_to_tensors=lists_to_tensors
+                )
+                num_verts_per_mesh = mesh.num_verts_per_mesh()
+                if force:
+                    # force mesh to have computed attributes
+                    mesh.verts_packed()
+                    mesh.edges_packed()
+                    mesh.laplacian_packed()
+                    mesh.faces_areas_packed()
+
+                new_verts = torch.rand((mesh._N, mesh._V, 3), device=mesh.device)
+                new_verts_list = [
+                    new_verts[i, : num_verts_per_mesh[i]] for i in range(N)
+                ]
+                new_mesh = mesh.update_padded(new_verts)
+
+                # check the attributes assigned at construction time
+                self.assertEqual(new_mesh._N, mesh._N)
+                self.assertEqual(new_mesh._F, mesh._F)
+                self.assertEqual(new_mesh._V, mesh._V)
+                self.assertEqual(new_mesh.equisized, mesh.equisized)
+                self.assertTrue(all(new_mesh.valid == mesh.valid))
+                self.assertNotSeparate(
+                    new_mesh.num_verts_per_mesh(), mesh.num_verts_per_mesh()
+                )
+                self.assertClose(
+                    new_mesh.num_verts_per_mesh(), mesh.num_verts_per_mesh()
+                )
+                self.assertNotSeparate(
+                    new_mesh.num_faces_per_mesh(), mesh.num_faces_per_mesh()
+                )
+                self.assertClose(
+                    new_mesh.num_faces_per_mesh(), mesh.num_faces_per_mesh()
+                )
+
+                # check that the following attributes are not assigned
+                self.assertIsNone(new_mesh._verts_list)
+                self.assertIsNone(new_mesh._faces_areas_packed)
+                self.assertIsNone(new_mesh._faces_normals_packed)
+                self.assertIsNone(new_mesh._verts_normals_packed)
+
+                check_tensors = [
+                    "_faces_packed",
+                    "_verts_packed_to_mesh_idx",
+                    "_faces_packed_to_mesh_idx",
+                    "_mesh_to_verts_packed_first_idx",
+                    "_mesh_to_faces_packed_first_idx",
+                    "_edges_packed",
+                    "_edges_packed_to_mesh_idx",
+                    "_mesh_to_edges_packed_first_idx",
+                    "_faces_packed_to_edges_packed",
+                    "_num_edges_per_mesh",
+                ]
+                for k in check_tensors:
+                    v = getattr(new_mesh, k)
+                    if not force:
+                        self.assertIsNone(v)
+                    else:
+                        v_old = getattr(mesh, k)
+                        self.assertNotSeparate(v, v_old)
+                        self.assertClose(v, v_old)
+
+                # check verts/faces padded
+                self.assertClose(new_mesh.verts_padded(), new_verts)
+                self.assertNotSeparate(new_mesh.verts_padded(), new_verts)
+                self.assertClose(new_mesh.faces_padded(), mesh.faces_padded())
+                self.assertNotSeparate(new_mesh.faces_padded(), mesh.faces_padded())
+                # check verts/faces list
+                for i in range(N):
+                    self.assertNotSeparate(
+                        new_mesh.faces_list()[i], mesh.faces_list()[i]
+                    )
+                    self.assertClose(new_mesh.faces_list()[i], mesh.faces_list()[i])
+                    self.assertSeparate(new_mesh.verts_list()[i], mesh.verts_list()[i])
+                    self.assertClose(new_mesh.verts_list()[i], new_verts_list[i])
+                # check verts/faces packed
+                self.assertClose(new_mesh.verts_packed(), torch.cat(new_verts_list))
+                self.assertSeparate(new_mesh.verts_packed(), mesh.verts_packed())
+                self.assertClose(new_mesh.faces_packed(), mesh.faces_packed())
+                # check pad_to_packed
+                self.assertClose(
+                    new_mesh.verts_padded_to_packed_idx(),
+                    mesh.verts_padded_to_packed_idx(),
+                )
+                # check edges
+                self.assertClose(new_mesh.edges_packed(), mesh.edges_packed())
+
     def test_get_mesh_verts_faces(self):
         device = torch.device("cuda:0")
         verts_list = []
