CtrlK
BlogDocsLog inGet started
Tessl Logo

tessl/pypi-rembg

Remove image background using advanced AI models including U-Net, BiRefNet, and SAM with support for multiple input formats and GPU acceleration

84

0.94x
Overview
Eval results
Files

rubric.jsonevals/scenario-5/

{
  "context": "This criteria evaluates how well the engineer uses the rembg package's post-processing capability for mask smoothing to improve the quality of background removal results. The focus is on correct usage of the post_process_mask parameter and proper application of the remove() function.",
  "type": "weighted_checklist",
  "checklist": [
    {
      "name": "Import rembg",
      "description": "Correctly imports the remove function from the rembg package (e.g., 'from rembg import remove')",
      "max_score": 10
    },
    {
      "name": "Basic remove usage",
      "description": "Uses the rembg remove() function to process images for background removal in the remove_background() implementation",
      "max_score": 20
    },
    {
      "name": "File I/O handling",
      "description": "Correctly reads input images from file paths and writes output images to file paths using appropriate methods (e.g., open with 'rb' mode for reading, write bytes or PIL Image save for output)",
      "max_score": 15
    },
    {
      "name": "post_process_mask parameter",
      "description": "Correctly uses the post_process_mask parameter when calling remove(), setting it to True when smooth_mask argument is True",
      "max_score": 30
    },
    {
      "name": "Standard vs smoothed",
      "description": "Implements compare_removal_quality() to call remove() twice on the same image: once without post_process_mask (or False) and once with post_process_mask=True",
      "max_score": 25
    }
  ]
}

Install with Tessl CLI

npx tessl i tessl/pypi-rembg

tile.json