Rotobot Next CLI Reference

CLI Arguments

	usage: rotobot_next [-h] [--mattes_folder MATTES_FOLDER] --mask_type {sam3,vit,user} --image_folder IMAGE_FOLDER --depth_folder DEPTH_FOLDER [--output_folder OUTPUT_FOLDER] [--checkpoint_path CHECKPOINT_PATH] [--detector_name DETECTOR_NAME]
                    [--segmentor_name SEGMENTOR_NAME] [--fov_name FOV_NAME] [--detector_path DETECTOR_PATH] [--segmentor_path SEGMENTOR_PATH] [--fov_path FOV_PATH] [--mhr_path MHR_PATH] [--bbox_thresh BBOX_THRESH] [--use_mask] [--debug_output]
	                [--mattes_output] [--filled_shapes_per_person_output] [--z_threshold Z_THRESHOLD]
	
	Rotobot Next from Tokgan: Creates visualisation and JSON file structure for spline based mattes                                     
	                                                                                                                                    
	options:                                                                                                                            
	  -h, --help            show this help message and exit                                                                             
	  --mattes_folder MATTES_FOLDER                                                                                                     
	                        Path to folder of matte images                                                                              
	  --mask_type {sam3,vit,user}                                                                                                       
	                        Path to file string of matte images                                                                         
	  --image_folder IMAGE_FOLDER                                                                                                       
	                        Path to folder containing input images                                                                      
	  --depth_folder DEPTH_FOLDER                                                                                                       
	                        Path to folder containing input images                                                                      
	  --output_folder OUTPUT_FOLDER                                                                                                     
	                        Path to output folder (default: ./output/)                                               
	  --checkpoint_path CHECKPOINT_PATH                                                                                                 
	                        Path to SAM 3D Body model checkpoint                                                                        
	  --detector_name DETECTOR_NAME                                                                                                     
	                        Human detection model for demo (Default `vitdet`, add your favorite detector if needed).                    
	  --segmentor_name SEGMENTOR_NAME                                                                                                   
	                        Human segmentation model for demo (Default `sam2`, add your favorite segmentor if needed).                  
	  --fov_name FOV_NAME   FOV estimation model for demo (Default `moge2`, add your favorite fov estimator if needed).                 
	  --detector_path DETECTOR_PATH                                                                                                     
	                        Path to human detection model folder (or set SAM3D_DETECTOR_PATH)                                           
	  --segmentor_path SEGMENTOR_PATH                                                                                                   
	                        Path to human segmentation model folder (or set SAM3D_SEGMENTOR_PATH)                                       
	  --fov_path FOV_PATH   Path to fov estimation model folder (or set SAM3D_FOV_PATH)                                                 
	  --mhr_path MHR_PATH   Path to MoHR/assets folder (or set SAM3D_mhr_path)                                                          
	  --bbox_thresh BBOX_THRESH                                                                                                         
	                        Bounding box detection threshold                                                                            
	  --use_mask            Use mask-conditioned prediction (segmentation mask is automatically generated from bbox)                    
	  --debug_output        Create pixel based debug                                                                                    
	  --mattes_output       Create pixel based mattes                                                                                   
	  --filled_shapes_per_person_output                                                                                                 
	                        Create pixel based mattes                                                                                   
	  --z_threshold Z_THRESHOLD                                                                                                         
	                        depth threshold in metres gets multipled by body thickness                                                  
	                                                                                                                                    
	                Examples:                                                                                                           
	                rotobot_next --image_folder ./images --mask_type vit --output_folder ./json                                          
	                                                                                                                                      
	                Environment Variables:                                                                                                
	                SAM3D_MHR_PATH: Path to MHR asset                                                                                     
	                SAM3D_DETECTOR_PATH: Path to human detection model folder                                                             
	                SAM3D_FOV_PATH: Path to fov estimation model folder                                                                 
	                SAM3D_SEGMENTOR_PATH: Path to human segmentation model folder                                                        
	                SAM3D_FOV_PATH: Path to fov estimation model folder                                                                 
	
	

Dependencies & Open Source

The following libraries are packaged and utilized within the Rotobot Next engine: