1. Feature Introduction

In Deep Learning models, Shadow Mode is a data enhancement method that reuses real data accumulated during model operation in actual scenes for model training, increasing data diversity and improving the model's generalization capability.

2. Applicable Scenarios

When the imaging model currently used by the Camera produces good imaging in 70%~90% of the area (but there is local collapse in the workpiece/container Point Cloud), and you need to continuously improve the Point Cloud imaging effect, you can use real-scene shadow data for training to obtain a fine-tuned model with better performance (good imaging in 95%+ of the area).

3. Steps for Use

（1）Check whether Shadow Mode is enabled.

1. Make sure the total amount of shadow data meets the requirement, with at least 20 shadow data samples [only data saved when triggered by the Robot arm]

   

   

2. On the main interface, click Camera**，enter the Camera configuration interface to connect the binocular Camera,**scroll down to Shadow Mode

3. Keep the number of iterations at the default value

4. Set the model name

5. Set remarks (optional)

   

6. Click Start Training and wait for the model training to complete

7. After training is completed, you can select the corresponding new model in model configuration and click Trigger Capture to view the effect.

   

Previous pageHow to Use Target Object Shadow Mode

Next pageHow to Use Mask Mode