Zed Camera Improvements

[Ishaan-Datta]

Combining depth sensing with OpenCV color filter for even more accurate object detection

• Runtime Parameters
• Depth Mode
• ROI
• Camera Parameters
• Camera Object
• View Field
• Init Parameters
• Device Parameters
• Sensors Module
• Depth Sensing
• Positional Tracking
• Object Group
• Custom Object Detection
• Video API

Tracking:

• Using pre-developed euclidean tracker

General Ideas:

• Zed Custom Detection Depth
• Apply Kalman filtering techniques to estimate the temporal alignment between RGB and depth frames. Utilize the predicted alignment to adjust the timing of one of the streams for synchronization.
• Use compression standards like JPEG or PNG for RGB images and compression algorithms like run-length encoding or delta encoding for depth data.
• Utilize techniques like blob detection, edge detection, or texture analysis to extract meaningful features.
• Normalize depth values to a consistent range to improve comparability between frames and scenes. Techniques such as min-max normalization or z-score normalization can be used depending on the application.
• Experiment with using depth sensor data for foreground isolation within OpenCV
• Apply appropriate filters to reduce noise in both RGB and depth images. Common filters include Gaussian or median filters.
• Consider using adaptive filtering techniques that adjust filter parameters based on local image characteristics.
• Perform depth map enhancement techniques to improve depth perception, such as edge-preserving smoothing or hole filling algorithms.
• Consider techniques like bilateral filtering or guided filtering to preserve depth edges while smoothing.
• Ensure proper alignment of RGB and depth images through calibration and registration processes.
• Use techniques like feature-based registration or intensity-based registration to align the two modalities accurately.