fix: resolve tensor type mismatch in depth estimation

- Fixed tensor type mismatch error between DoubleTensor and FloatTensor
- Added explicit tensor type conversion in MiDaSWrapper
- Improved error handling for tensor type mismatches
- Enhanced process_image to handle float64 tensors properly
- Added debugging logs to track tensor types
This commit is contained in:
limbicnation
2025-05-03 23:32:15 +02:00
parent 0438e413ef
commit 2914dd8b5c
2 changed files with 884 additions and 58 deletions
+877 -56
View File
File diff suppressed because it is too large Load Diff
+7 -2
View File
@@ -15,8 +15,13 @@ Pillow>=9.2.0 # Ensures compatibility with nodes requiring >=9.2.0
numpy>=1.23.0 # Resolves compatibility issues
# Additional dependencies specific to depth estimation node
timm>=0.6.12 # Required for depth estimation models
timm>=0.6.12 # Required for Depth Anything models
huggingface-hub>=0.16.0 # For model downloading
wget>=3.2 # For reliable model downloading
# Torch version requirements
# Note: PyTorch dependencies are handled by ComfyUI's core installation
# If you're installing this node directly, ensure torch>=2.0.0 is available
# If you're installing this node directly, ensure torch>=2.0.0 is available
# Network dependencies
requests>=2.27.0 # For model downloading