Upload pytorch3d_stub/pytorch3d/renderer/__init__.py with huggingface_hub

pytorch3d_stub/pytorch3d/renderer/__init__.py

@@ -0,0 +1,48 @@
+"""pytorch3d.renderer stub – look_at_view_transform only."""
+import torch
+import math
+
+def look_at_view_transform(dist=1.0, elev=0.0, azim=0.0, degrees=True, eye=None,
+                            at=((0, 0, 0),), up=((0, 1, 0),), device="cpu"):
+    """Compute R, T for look-at camera transform (pytorch3d convention)."""
+    if eye is not None:
+        if not isinstance(eye, torch.Tensor):
+            eye = torch.tensor(eye, dtype=torch.float32, device=device)
+        if eye.dim() == 1:
+            eye = eye.unsqueeze(0)
+    else:
+        if degrees:
+            elev_r = math.radians(elev) if isinstance(elev, (int, float)) else torch.deg2rad(torch.tensor(elev, dtype=torch.float32))
+            azim_r = math.radians(azim) if isinstance(azim, (int, float)) else torch.deg2rad(torch.tensor(azim, dtype=torch.float32))
+        else:
+            elev_r = elev
+            azim_r = azim
+        if isinstance(elev_r, (int, float)):
+            x = dist * math.cos(elev_r) * math.sin(azim_r)
+            y = dist * math.sin(elev_r)
+            z = dist * math.cos(elev_r) * math.cos(azim_r)
+            eye = torch.tensor([[x, y, z]], dtype=torch.float32, device=device)
+        else:
+            x = dist * torch.cos(elev_r) * torch.sin(azim_r)
+            y = dist * torch.sin(elev_r)
+            z = dist * torch.cos(elev_r) * torch.cos(azim_r)
+            eye = torch.stack([x, y, z], dim=-1).unsqueeze(0).to(device)
+
+    if not isinstance(at, torch.Tensor):
+        at = torch.tensor(at, dtype=torch.float32, device=device)
+    if at.dim() == 1:
+        at = at.unsqueeze(0)
+    if not isinstance(up, torch.Tensor):
+        up = torch.tensor(up, dtype=torch.float32, device=device)
+    if up.dim() == 1:
+        up = up.unsqueeze(0)
+
+    z_axis = eye - at
+    z_axis = z_axis / z_axis.norm(dim=-1, keepdim=True).clamp(min=1e-8)
+    x_axis = torch.cross(up.expand_as(z_axis), z_axis, dim=-1)
+    x_axis = x_axis / x_axis.norm(dim=-1, keepdim=True).clamp(min=1e-8)
+    y_axis = torch.cross(z_axis, x_axis, dim=-1)
+
+    R = torch.stack([x_axis, y_axis, z_axis], dim=-1)  # (N, 3, 3)
+    T = -torch.bmm(R.transpose(1, 2), eye.unsqueeze(-1)).squeeze(-1)  # (N, 3)
+    return R, T