camera module #1
Conversation
src/humanoid_vision/camera.py
Outdated
| raise NotImplementedError("Camera module not yet implemented") | ||
|
|
||
| MAX_IDX = 10 | ||
| #cv2.utils.logging.setLogLevel(cv2.utils.logging.LOG_LEVEL_ERROR) |
There was a problem hiding this comment.
Remove the commented out lines if possible
src/humanoid_vision/camera.py
Outdated
| import torch | ||
| import torch.nn as nn | ||
|
|
||
| def placeholder_camera() -> Any: |
There was a problem hiding this comment.
You can get rid of the placeholder now
src/humanoid_vision/camera.py
Outdated
| print(available) | ||
|
|
||
| class Camera(): | ||
| def __init__(self, idx: int): #could also be a string |
There was a problem hiding this comment.
What is the difference if it is an int vs a string?
src/humanoid_vision/camera.py
Outdated
| def set_brightness(self, val:float): | ||
| self.cam.set(cv2.CAP_PROP_BRIGHTNESS, val) | ||
|
|
||
| def test(): |
There was a problem hiding this comment.
Could you refactor the tests to use pytest, and move it to a tests folder?
There was a problem hiding this comment.
Pull Request Overview
This PR implements a basic camera module for humanoid vision functionality, replacing a placeholder implementation with OpenCV-based camera access. The module provides camera device discovery, initialization, and basic property control.
- Adds camera device enumeration and availability checking
- Implements a Camera class with initialization and brightness control
- Includes a test function to demonstrate camera functionality
Tip: Customize your code reviews with copilot-instructions.md. Create the file or learn how to get started.
src/humanoid_vision/camera.py
Outdated
| from typing import Any | ||
|
|
||
| import cv2 | ||
| from cv2_enumerate_cameras import enumerate_cameras |
There was a problem hiding this comment.
The cv2_enumerate_cameras import is unused in the code. Consider removing it or using it to replace the manual camera enumeration loop.
Copilot uses AI. Check for mistakes.
src/humanoid_vision/camera.py
Outdated
| import torch | ||
| import torch.nn as nn |
There was a problem hiding this comment.
The torch imports are unused in this module. Remove these unused imports to keep the code clean.
| import torch | |
| import torch.nn as nn |
Copilot uses AI. Check for mistakes.
| available = [] | ||
| for i in range(MAX_IDX): | ||
| cap = cv2.VideoCapture(i) | ||
| if cap.isOpened(): | ||
| available.append(i) | ||
| cap.release() | ||
|
|
||
| print(available) |
There was a problem hiding this comment.
Global camera enumeration code should be moved inside a function or class method. Module-level execution can cause issues during imports and testing.
Copilot uses AI. Check for mistakes.
| available.append(i) | ||
| cap.release() | ||
|
|
||
| print(available) |
There was a problem hiding this comment.
Remove debug print statement from production code. Consider using logging instead if this information is needed.
Copilot uses AI. Check for mistakes.
src/humanoid_vision/camera.py
Outdated
| raise ValueError(f"id isn't an available device, select from {available}") | ||
|
|
||
| self.cam = cv2.VideoCapture(idx) | ||
| # Show the frame in a window |
There was a problem hiding this comment.
Remove orphaned comment that doesn't match the actual code. This comment appears to be leftover from development.
| # Show the frame in a window |
Copilot uses AI. Check for mistakes.
src/humanoid_vision/camera.py
Outdated
|
|
||
| print(available) | ||
|
|
||
| class Camera(): |
There was a problem hiding this comment.
Remove empty parentheses after class name. In Python, class Camera: is preferred over class Camera():.
| class Camera(): | |
| class Camera: |
Copilot uses AI. Check for mistakes.
src/humanoid_vision/camera.py
Outdated
| class Camera(): | ||
| def __init__(self, idx: int): #could also be a string | ||
| """ | ||
| initialize instance to a cv2 index |
There was a problem hiding this comment.
Capitalize the first word of the docstring: 'Initialize instance to a cv2 index'.
| initialize instance to a cv2 index | |
| Initialize instance to a cv2 index |
Copilot uses AI. Check for mistakes.
2 participants
approve me twin