block.py

from typing import List, Union

import av
import numpy as np
import torch
from diffusers.modular_pipelines import (
    ComponentSpec,
    InputParam,
    ModularPipelineBlocks,
    OutputParam,
    PipelineState,
)
from matplotlib import colormaps
from PIL import Image
from transformers import DepthAnythingForDepthEstimation, DPTImageProcessor

TURBO_CMAP = colormaps["turbo"]


def save_video(frames: List[Image.Image], fps: float, output_path: str) -> None:
    """Save a list of PIL Image frames as an MP4 video."""
    container = av.open(output_path, mode="w")
    stream = container.add_stream("libx264", rate=int(fps))
    stream.pix_fmt = "yuv420p"
    stream.width = frames[0].width
    stream.height = frames[0].height

    for frame in frames:
        video_frame = av.VideoFrame.from_image(frame)
        for packet in stream.encode(video_frame):
            container.mux(packet)

    for packet in stream.encode():
        container.mux(packet)
    container.close()


class DepthAnythingV2EstimatorBlock(ModularPipelineBlocks):
    _requirements = {
        "transformers": ">=5.1.0",
        "torch": ">=2.9.0",
        "torchvision": ">=0.16.0",
        "av": ">=12.0.0",
        "matplotlib": ">=3.7.0",
    }

    @property
    def expected_components(self) -> List[ComponentSpec]:
        return [
            ComponentSpec(
                name="depth_estimator",
                type_hint=DepthAnythingForDepthEstimation,
                pretrained_model_name_or_path="depth-anything/Depth-Anything-V2-Large-hf",
            ),
            ComponentSpec(
                name="depth_estimator_processor",
                type_hint=DPTImageProcessor,
                pretrained_model_name_or_path="depth-anything/Depth-Anything-V2-Large-hf",
            ),
        ]

    @property
    def inputs(self) -> List[InputParam]:
        return [
            InputParam(
                "image",
                type_hint=Union[Image.Image, List[Image.Image]],
                required=False,
                description="Image(s) to estimate depth for",
            ),
            InputParam(
                "video_path",
                type_hint=str,
                required=False,
                description="Path to input video file. When provided, image is ignored.",
            ),
            InputParam(
                "colormap",
                type_hint=str,
                default="grayscale",
                description="Depth visualization format: 'grayscale' or 'turbo' (colormapped)",
            ),
        ]

    @property
    def intermediate_outputs(self) -> List[OutputParam]:
        return [
            OutputParam(
                "depth_image",
                type_hint=Image.Image,
                description="Normalized depth map as a PIL image (single image mode)",
            ),
            OutputParam(
                "predicted_depth",
                type_hint=torch.Tensor,
                description="Raw relative depth tensor (H x W) (single image mode)",
            ),
            OutputParam(
                "depth_frames",
                type_hint=list,
                description="List of per-frame depth PIL images (video mode)",
            ),
            OutputParam(
                "fps",
                type_hint=float,
                description="Source video frame rate (video mode)",
            ),
        ]

    def _estimate_depth(self, image: Image.Image, processor, model) -> dict:
        inputs = processor(images=[image], return_tensors="pt").to(model.device)
        outputs = model(**inputs)
        post_processed = processor.post_process_depth_estimation(
            outputs, target_sizes=[(image.height, image.width)]
        )
        return post_processed[0]

    def _normalize_depth(self, depth: np.ndarray) -> np.ndarray:
        d_min = depth.min()
        d_max = depth.max()
        normalized = (depth - d_min) / (d_max - d_min + 1e-8)
        # Invert so bright = close, dark = far (consistent with depth_pro)
        return 1.0 - normalized

    def _apply_colormap(self, normalized: np.ndarray, mode: str) -> np.ndarray:
        if mode == "turbo":
            colored = (TURBO_CMAP(normalized)[..., :3] * 255).astype(np.uint8)
            return colored
        return (normalized * 255.0).astype(np.uint8)

    def _process_video(self, video_path, processor, model, colormap):
        input_container = av.open(video_path)
        video_stream = input_container.streams.video[0]
        fps = video_stream.average_rate

        depth_frames = []
        for frame in input_container.decode(video=0):
            pil_image = frame.to_image().convert("RGB")

            result = self._estimate_depth(pil_image, processor, model)
            depth_np = result["predicted_depth"].float().cpu().numpy()
            normalized = self._normalize_depth(depth_np)
            colored = self._apply_colormap(normalized, colormap)

            if colormap == "turbo":
                depth_frame = Image.fromarray(colored, mode="RGB")
            else:
                depth_frame = Image.fromarray(colored, mode="L")
            depth_frames.append(depth_frame)

        input_container.close()

        return depth_frames, fps

    @torch.no_grad()
    def __call__(self, components, state: PipelineState) -> PipelineState:
        block_state = self.get_block_state(state)

        processor = components.depth_estimator_processor
        model = components.depth_estimator

        video_path = getattr(block_state, "video_path", None)

        if video_path:
            depth_frames, fps = self._process_video(
                video_path, processor, model, block_state.colormap
            )
            block_state.depth_frames = depth_frames
            block_state.fps = float(fps)
            block_state.depth_image = None
            block_state.predicted_depth = None
        else:
            image = block_state.image
            if not isinstance(image, list):
                image = [image]

            result = self._estimate_depth(image[0], processor, model)
            predicted_depth = result["predicted_depth"]

            block_state.predicted_depth = predicted_depth

            depth_np = predicted_depth.float().cpu().numpy()
            normalized = self._normalize_depth(depth_np)
            colored = self._apply_colormap(normalized, block_state.colormap)
            if block_state.colormap == "turbo":
                block_state.depth_image = Image.fromarray(colored, mode="RGB")
            else:
                block_state.depth_image = Image.fromarray(colored, mode="L")

            block_state.depth_frames = None
            block_state.fps = None

        self.set_block_state(state, block_state)

        return components, state