Description #

Image aesthetic scoring processor, intelligently evaluates the aesthetic quality and composition of images

Key features #

• Aesthetic scoring: Professional evaluation of the aesthetic quality of input images
• Composition analysis: Analyzes image composition based on visual perception theory
• Standardized output: Scores are normalized to the range 0–1 for easier subsequent processing
• Batch processing: Supports efficient batch image scoring
• Multi-format support: Compatible with multiple image input formats

Format support #

• Input: Supports multiple formats including image URLs, TOS addresses, and binary streams
• Output: Floating-point aesthetic score (0.0–1.0); higher values indicate better aesthetic quality
• Image formats: Mainstream formats such as JPG, PNG, WebP
• Resolution: Automatically adapts to different image input resolutions

Scoring principles #

• Visual feature extraction: Uses advanced visual models to extract image features
• Aesthetic modeling: Scoring model trained on large-scale aesthetic datasets
• Multidimensional evaluation: Considers multiple aesthetic elements including color, composition, and contrast
• Perceptual alignment: Scoring results are highly consistent with human aesthetic perception

Operator parameters #

Input #

Output #

Floating-point array containing aesthetic scores, score range 0.0–1.0,
Images that cannot be decoded or fail to process return a null value

Parameters #

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

Examples #

The following code demonstrates how to use Daft (for distributed environments) to run the operator and score the aesthetic quality of images.

from __future__ import annotations

import os

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

if __name__ == "__main__":
    TOS_TEST_DIR_URL = os.getenv("TOS_TEST_DIR_URL", "las-cn-beijing-public-online.tos-cn-beijing.volces.com")
    model_path = os.getenv("MODEL_PATH", "./models")

    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",
            )
            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)

    samples = {
        "input_path": [f"https://{TOS_TEST_DIR_URL}/public/archive/image_aesthetic_score/forest.jpg"],
    }
    ds = daft.from_pydict(samples)

    # Using Daft to calculate aesthetic scores for images
    constructor_kwargs = {
        "model_path": model_path,
        "image_src_type": "image_url",
        "batch_size": 1,
    }

    ds = ds.with_column(
        "aesthetic_score",
        las_udf(
            ImageAestheticScore,
            construct_args=constructor_kwargs,
            num_cpus=1,
            concurrency=1,
            batch_size=1,
        )(col("input_path")),
    )

    ds.show()
    # ╭────────────────────────────────┬────────────────────╮
    # │ input_path                     ┆ aesthetic_score    │
    # │ ---                            ┆ ---                │
    # │ Utf8                           ┆ Float64            │
    # ╞════════════════════════════════╪════════════════════╡
    # │ https://las-public-data-qa.to… ┆ 0.5603389739990234 │
    # ╰────────────────────────────────┴────────────────────╯