deploy: fc0c10e734116107123b6dce81a6df2cbbf84dfe

layout-occlusion.py

@@ -0,0 +1,147 @@
+import os
+from typing import List, Union
+
+import datasets as ds
+import evaluate
+import numpy as np
+import numpy.typing as npt
+from PIL import Image
+
+_DESCRIPTION = r"""\
+Computes the average pixel value of areas covered by elements in S.
+"""
+
+_KWARGS_DESCRIPTION = """\
+FIXME
+"""
+
+_CITATION = """\
+@inproceedings{hsu2023posterlayout,
+  title={Posterlayout: A new benchmark and approach for content-aware visual-textual presentation layout},
+  author={Hsu, Hsiao Yuan and He, Xiangteng and Peng, Yuxin and Kong, Hao and Zhang, Qing},
+  booktitle={Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition},
+  pages={6018--6026},
+  year={2023}
+}
+"""
+
+
+class LayoutOcculusion(evaluate.Metric):
+    def __init__(
+        self,
+        canvas_width: int,
+        canvas_height: int,
+        **kwargs,
+    ) -> None:
+        super().__init__(**kwargs)
+        self.canvas_width = canvas_width
+        self.canvas_height = canvas_height
+
+    def _info(self) -> evaluate.EvaluationModuleInfo:
+        return evaluate.MetricInfo(
+            description=_DESCRIPTION,
+            citation=_CITATION,
+            inputs_description=_KWARGS_DESCRIPTION,
+            features=ds.Features(
+                {
+                    "predictions": ds.Sequence(ds.Sequence(ds.Value("float64"))),
+                    "gold_labels": ds.Sequence(ds.Sequence(ds.Value("int64"))),
+                    "saliency_maps_1": ds.Sequence(ds.Value("string")),
+                    "saliency_maps_2": ds.Sequence(ds.Value("string")),
+                }
+            ),
+            codebase_urls=[
+                "https://github.com/PKU-ICST-MIPL/PosterLayout-CVPR2023/blob/main/eval.py#L144-L171"
+            ],
+        )
+
+    def load_saliency_map(
+        self,
+        filepath: Union[os.PathLike, List[os.PathLike]],
+    ) -> npt.NDArray[np.float64]:
+        if isinstance(filepath, list):
+            assert len(filepath) == 1, filepath
+            filepath = filepath[0]
+
+        map_pil = Image.open(filepath)  # type: ignore
+        map_pil = map_pil.convert("L")
+
+        if map_pil.size != (self.canvas_width, self.canvas_height):
+            map_pil = map_pil.resize((self.canvas_width, self.canvas_height))
+
+        map_arr = np.array(map_pil)
+        map_arr = map_arr / 255.0
+        return map_arr
+
+    def get_rid_of_invalid(
+        self, predictions: npt.NDArray[np.float64], gold_labels: npt.NDArray[np.int64]
+    ) -> npt.NDArray[np.int64]:
+        assert len(predictions) == len(gold_labels)
+
+        w = self.canvas_width / 100
+        h = self.canvas_height / 100
+
+        for i, prediction in enumerate(predictions):
+            for j, b in enumerate(prediction):
+                xl, yl, xr, yr = b
+                xl = max(0, xl)
+                yl = max(0, yl)
+                xr = min(self.canvas_width, xr)
+                yr = min(self.canvas_height, yr)
+                if abs((xr - xl) * (yr - yl)) < w * h * 10:
+                    if gold_labels[i, j]:
+                        gold_labels[i, j] = 0
+        return gold_labels
+
+    def _compute(
+        self,
+        *,
+        predictions: Union[npt.NDArray[np.float64], List[List[float]]],
+        gold_labels: Union[npt.NDArray[np.int64], List[int]],
+        saliency_maps_1: List[os.PathLike],
+        saliency_maps_2: List[os.PathLike],
+    ) -> float:
+        predictions = np.array(predictions)
+        gold_labels = np.array(gold_labels)
+
+        predictions[:, :, ::2] *= self.canvas_width
+        predictions[:, :, 1::2] *= self.canvas_height
+
+        gold_labels = self.get_rid_of_invalid(
+            predictions=predictions, gold_labels=gold_labels
+        )
+
+        score = 0.0
+
+        assert (
+            len(predictions)
+            == len(gold_labels)
+            == len(saliency_maps_1)
+            == len(saliency_maps_2)
+        )
+        num_predictions = len(predictions)
+        it = zip(predictions, gold_labels, saliency_maps_1, saliency_maps_2)
+
+        for prediction, gold_label, smap_1, smap_2 in it:
+            smap_arr_1 = self.load_saliency_map(smap_1)
+            smap_arr_2 = self.load_saliency_map(smap_2)
+
+            smap_arr = np.maximum(smap_arr_1, smap_arr_2)
+            cal_mask = np.zeros_like(smap_arr)
+
+            prediction = np.array(prediction, dtype=int)
+            gold_label = np.array(gold_label, dtype=int)
+
+            mask = (gold_label > 0).reshape(-1)
+            mask_prediction = prediction[mask]
+
+            for mp in mask_prediction:
+                xl, yl, xr, yr = mp
+                cal_mask[yl:yr, xl:xr] = 1
+
+            total_area = np.sum(cal_mask)
+            total_sal = np.sum(smap_arr[cal_mask == 1])
+            if total_sal and total_area:
+                score += total_sal / total_area
+
+        return score / num_predictions

requirements.txt

@@ -0,0 +1,89 @@
+aiofiles==23.2.1 ; python_version >= "3.9" and python_version < "4.0"
+aiohttp==3.9.3 ; python_version >= "3.9" and python_version < "4.0"
+aiosignal==1.3.1 ; python_version >= "3.9" and python_version < "4.0"
+altair==5.2.0 ; python_version >= "3.9" and python_version < "4.0"
+annotated-types==0.6.0 ; python_version >= "3.9" and python_version < "4.0"
+anyio==4.2.0 ; python_version >= "3.9" and python_version < "4.0"
+arrow==1.3.0 ; python_version >= "3.9" and python_version < "4.0"
+async-timeout==4.0.3 ; python_version >= "3.9" and python_version < "3.11"
+attrs==23.2.0 ; python_version >= "3.9" and python_version < "4.0"
+binaryornot==0.4.4 ; python_version >= "3.9" and python_version < "4.0"
+certifi==2024.2.2 ; python_version >= "3.9" and python_version < "4.0"
+chardet==5.2.0 ; python_version >= "3.9" and python_version < "4.0"
+charset-normalizer==3.3.2 ; python_version >= "3.9" and python_version < "4.0"
+click==8.1.7 ; python_version >= "3.9" and python_version < "4.0"
+colorama==0.4.6 ; python_version >= "3.9" and python_version < "4.0"
+contourpy==1.2.0 ; python_version >= "3.9" and python_version < "4.0"
+cookiecutter==2.5.0 ; python_version >= "3.9" and python_version < "4.0"
+cycler==0.12.1 ; python_version >= "3.9" and python_version < "4.0"
+datasets==2.17.0 ; python_version >= "3.9" and python_version < "4.0"
+dill==0.3.8 ; python_version >= "3.9" and python_version < "4.0"
+evaluate[template]==0.4.1 ; python_version >= "3.9" and python_version < "4.0"
+exceptiongroup==1.2.0 ; python_version >= "3.9" and python_version < "3.11"
+fastapi==0.109.2 ; python_version >= "3.9" and python_version < "4.0"
+ffmpy==0.3.1 ; python_version >= "3.9" and python_version < "4.0"
+filelock==3.13.1 ; python_version >= "3.9" and python_version < "4.0"
+fonttools==4.48.1 ; python_version >= "3.9" and python_version < "4.0"
+frozenlist==1.4.1 ; python_version >= "3.9" and python_version < "4.0"
+fsspec==2023.10.0 ; python_version >= "3.9" and python_version < "4.0"
+fsspec[http]==2023.10.0 ; python_version >= "3.9" and python_version < "4.0"
+gradio-client==0.10.0 ; python_version >= "3.9" and python_version < "4.0"
+gradio==4.18.0 ; python_version >= "3.9" and python_version < "4.0"
+h11==0.14.0 ; python_version >= "3.9" and python_version < "4.0"
+httpcore==1.0.2 ; python_version >= "3.9" and python_version < "4.0"
+httpx==0.26.0 ; python_version >= "3.9" and python_version < "4.0"
+huggingface-hub==0.20.3 ; python_version >= "3.9" and python_version < "4.0"
+idna==3.6 ; python_version >= "3.9" and python_version < "4.0"
+importlib-resources==6.1.1 ; python_version >= "3.9" and python_version < "4.0"
+jinja2==3.1.3 ; python_version >= "3.9" and python_version < "4.0"
+jsonschema-specifications==2023.12.1 ; python_version >= "3.9" and python_version < "4.0"
+jsonschema==4.21.1 ; python_version >= "3.9" and python_version < "4.0"
+kiwisolver==1.4.5 ; python_version >= "3.9" and python_version < "4.0"
+markdown-it-py==3.0.0 ; python_version >= "3.9" and python_version < "4.0"
+markupsafe==2.1.5 ; python_version >= "3.9" and python_version < "4.0"
+matplotlib==3.8.2 ; python_version >= "3.9" and python_version < "4.0"
+mdurl==0.1.2 ; python_version >= "3.9" and python_version < "4.0"
+multidict==6.0.5 ; python_version >= "3.9" and python_version < "4.0"
+multiprocess==0.70.16 ; python_version >= "3.9" and python_version < "4.0"
+numpy==1.26.4 ; python_version >= "3.9" and python_version < "4.0"
+orjson==3.9.13 ; python_version >= "3.9" and python_version < "4.0"
+packaging==23.2 ; python_version >= "3.9" and python_version < "4.0"
+pandas==2.2.0 ; python_version >= "3.9" and python_version < "4.0"
+pillow==10.2.0 ; python_version >= "3.9" and python_version < "4.0"
+pyarrow-hotfix==0.6 ; python_version >= "3.9" and python_version < "4.0"
+pyarrow==15.0.0 ; python_version >= "3.9" and python_version < "4.0"
+pydantic-core==2.16.2 ; python_version >= "3.9" and python_version < "4.0"
+pydantic==2.6.1 ; python_version >= "3.9" and python_version < "4.0"
+pydub==0.25.1 ; python_version >= "3.9" and python_version < "4.0"
+pygments==2.17.2 ; python_version >= "3.9" and python_version < "4.0"
+pyparsing==3.1.1 ; python_version >= "3.9" and python_version < "4.0"
+python-dateutil==2.8.2 ; python_version >= "3.9" and python_version < "4.0"
+python-multipart==0.0.9 ; python_version >= "3.9" and python_version < "4.0"
+python-slugify==8.0.4 ; python_version >= "3.9" and python_version < "4.0"
+pytz==2024.1 ; python_version >= "3.9" and python_version < "4.0"
+pyyaml==6.0.1 ; python_version >= "3.9" and python_version < "4.0"
+referencing==0.33.0 ; python_version >= "3.9" and python_version < "4.0"
+requests==2.31.0 ; python_version >= "3.9" and python_version < "4.0"
+responses==0.18.0 ; python_version >= "3.9" and python_version < "4.0"
+rich==13.7.0 ; python_version >= "3.9" and python_version < "4.0"
+rpds-py==0.17.1 ; python_version >= "3.9" and python_version < "4.0"
+ruff==0.2.1 ; python_version >= "3.9" and python_version < "4.0"
+semantic-version==2.10.0 ; python_version >= "3.9" and python_version < "4.0"
+shellingham==1.5.4 ; python_version >= "3.9" and python_version < "4.0"
+six==1.16.0 ; python_version >= "3.9" and python_version < "4.0"
+sniffio==1.3.0 ; python_version >= "3.9" and python_version < "4.0"
+starlette==0.36.3 ; python_version >= "3.9" and python_version < "4.0"
+text-unidecode==1.3 ; python_version >= "3.9" and python_version < "4.0"
+tomlkit==0.12.0 ; python_version >= "3.9" and python_version < "4.0"
+toolz==0.12.1 ; python_version >= "3.9" and python_version < "4.0"
+tqdm==4.66.2 ; python_version >= "3.9" and python_version < "4.0"
+typer[all]==0.9.0 ; python_version >= "3.9" and python_version < "4.0"
+types-python-dateutil==2.8.19.20240106 ; python_version >= "3.9" and python_version < "4.0"
+typing-extensions==4.9.0 ; python_version >= "3.9" and python_version < "4.0"
+tzdata==2024.1 ; python_version >= "3.9" and python_version < "4.0"
+urllib3==2.2.0 ; python_version >= "3.9" and python_version < "4.0"
+uvicorn==0.27.1 ; python_version >= "3.9" and python_version < "4.0"
+websockets==11.0.3 ; python_version >= "3.9" and python_version < "4.0"
+xxhash==3.4.1 ; python_version >= "3.9" and python_version < "4.0"
+yarl==1.9.4 ; python_version >= "3.9" and python_version < "4.0"
+zipp==3.17.0 ; python_version >= "3.9" and python_version < "3.10"