Description #

Image face blur processing operator

Main features #

• Automatically detects faces in images and applies blur processing
• Supports multiple blur types (mean, box, Gaussian)
• Supports multiple input methods, including URL, local path, Base64, and binary
• Optionally outputs the Base64 encoding of the blurred image and the path where it is saved
• Can run in both CPU and GPU environments

Applicable scenarios #

• Privacy compliance: Anonymize and blur faces in images before publishing
• Data preprocessing: Generate versions of training or evaluation datasets with blurred faces
• Content review/display: Automatically blur faces in images with sensitive faces

Caution #

• If processing fails for a single sample, the result for that item returns {"base64": None, "image_path": None}, and other samples in the same batch are not affected
• When output_dir is configured, the blurred images will be uploaded to the specified TOS directory or saved to the local directory

Operator parameters #

Input #

Output #

An array of processing results, where each element includes:

• base64 (str | None): Base64 encoding of the blurred image.
• image_path (str | None): Local or TOS path of the blurred image; if saving fails or saving is not configured, returns None.
• face_bounding_boxes (list[tuple[int, int, int, int]] | None): List of detected face bounding boxes (x, y, w, h); returns None if no faces are detected or processing fails.

Parameters #

If a parameter does not have a default value, it is required

Examples #

The following code demonstrates how to use daft to run the operator and blur faces in images.

from __future__ import annotations

import os

import daft
from daft import col
from daft.las.functions.image import ImageFaceBlur
from daft.las.functions.udf import las_udf

if __name__ == "__main__":
    
    if os.getenv("DAFT_RUNNER", "native") == "ray":
        import logging

        import ray

        def configure_logging():
            logging.basicConfig(
                level=logging.INFO,
                format="%(asctime)s - %(name)s - %(levelname)s - %(message)s",
                datefmt="%Y-%m-%d %H:%M:%S.%s".format(),
            )
            logging.getLogger("tracing.span").setLevel(logging.WARNING)
            logging.getLogger("daft_io.stats").setLevel(logging.WARNING)
            logging.getLogger("DaftStatisticsManager").setLevel(logging.WARNING)
            logging.getLogger("DaftFlotillaScheduler").setLevel(logging.WARNING)
            logging.getLogger("DaftFlotillaDispatcher").setLevel(logging.WARNING)

        ray.init(dashboard_host="0.0.0.0", runtime_env={"worker_process_setup_hook": configure_logging})
        daft.set_runner_ray()

    daft.set_execution_config(actor_udf_ready_timeout=600)
    daft.set_execution_config(min_cpu_per_task=0)

    tos_dir_url = os.getenv("TOS_DIR_URL", "las-cn-beijing-public-online.tos-cn-beijing.volces.com")
    samples = {"image_path": [f"https://{tos_dir_url}/public/shared_image_dataset/mengnalisa.png"]}
    ds = daft.from_pydict(samples)

    model_path = os.getenv("MODEL_PATH", "/opt/las/models")
    constructor_kwargs = {
        "model_path": model_path,
        "blur_type": "gaussian",
        "radius": 20.0,
        "return_base64": True
    }

    ds = ds.with_column(
        "results",
        las_udf(
            ImageFaceBlur,
            construct_args=constructor_kwargs,
            num_gpus=0,
            batch_size=1,
            concurrency=1,
        )(col("image_path")),
    )

    ds.show()

    # ╭────────────────────────────────┬────────────────────────────────────────────────────────────────────────────────────╮
    # │ image_path                     ┆ results                                                                            │
    # │ ---                            ┆ ---                                                                                │
    # │ String                         ┆ Struct[base64: String, image_path: String, face_bounding_boxes: List[List[Int32]]] │
    # ╞════════════════════════════════╪════════════════════════════════════════════════════════════════════════════════════╡
    # │https://las-cn-beijing-public…  ┆ {base64: iVBORw0KGgoAAAANSUhE…                                                     │
    # ╰────────────────────────────────┴────────────────────────────────────────────────────────────────────────────────────╯