Add initial commit

.gitattributes

@@ -0,0 +1,37 @@
+*.7z filter=lfs diff=lfs merge=lfs -text
+*.arrow filter=lfs diff=lfs merge=lfs -text
+*.bin filter=lfs diff=lfs merge=lfs -text
+*.bz2 filter=lfs diff=lfs merge=lfs -text
+*.ckpt filter=lfs diff=lfs merge=lfs -text
+*.ftz filter=lfs diff=lfs merge=lfs -text
+*.gz filter=lfs diff=lfs merge=lfs -text
+*.h5 filter=lfs diff=lfs merge=lfs -text
+*.joblib filter=lfs diff=lfs merge=lfs -text
+*.lfs.* filter=lfs diff=lfs merge=lfs -text
+*.mlmodel filter=lfs diff=lfs merge=lfs -text
+*.model filter=lfs diff=lfs merge=lfs -text
+*.msgpack filter=lfs diff=lfs merge=lfs -text
+*.npy filter=lfs diff=lfs merge=lfs -text
+*.npz filter=lfs diff=lfs merge=lfs -text
+*.onnx filter=lfs diff=lfs merge=lfs -text
+*.ot filter=lfs diff=lfs merge=lfs -text
+*.parquet filter=lfs diff=lfs merge=lfs -text
+*.pb filter=lfs diff=lfs merge=lfs -text
+*.pickle filter=lfs diff=lfs merge=lfs -text
+*.pkl filter=lfs diff=lfs merge=lfs -text
+*.pt filter=lfs diff=lfs merge=lfs -text
+*.pth filter=lfs diff=lfs merge=lfs -text
+*.rar filter=lfs diff=lfs merge=lfs -text
+*.safetensors filter=lfs diff=lfs merge=lfs -text
+saved_model/**/* filter=lfs diff=lfs merge=lfs -text
+*.tar.* filter=lfs diff=lfs merge=lfs -text
+*.tar filter=lfs diff=lfs merge=lfs -text
+*.tflite filter=lfs diff=lfs merge=lfs -text
+*.tgz filter=lfs diff=lfs merge=lfs -text
+*.wasm filter=lfs diff=lfs merge=lfs -text
+*.xz filter=lfs diff=lfs merge=lfs -text
+*.zip filter=lfs diff=lfs merge=lfs -text
+*.zst filter=lfs diff=lfs merge=lfs -text
+*tfevents* filter=lfs diff=lfs merge=lfs -text
+*.png filter=lfs diff=lfs merge=lfs -text
+*.pt2 filter=lfs diff=lfs merge=lfs -text

README.md

@@ -0,0 +1,13 @@
+---
+title: Sapiens Segmentation
+emoji: 🌍
+colorFrom: blue
+colorTo: purple
+sdk: gradio
+sdk_version: 4.42.0
+app_file: app.py
+pinned: false
+license: cc-by-nc-4.0
+---
+
+Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference

app.py

@@ -0,0 +1,307 @@
+import os
+import cv2
+import gradio as gr
+import numpy as np
+import spaces
+import torch
+import torch.nn.functional as F
+from gradio.themes.utils import sizes
+from torchvision import transforms
+from PIL import Image
+import tempfile
+from classes_and_palettes import GOLIATH_PALETTE, GOLIATH_CLASSES
+
+class Config:
+    ASSETS_DIR = os.path.join(os.path.dirname(__file__), 'assets')
+    CHECKPOINTS_DIR = os.path.join(ASSETS_DIR, "checkpoints")
+    CHECKPOINTS = {
+        "0.3b": "sapiens_0.3b_goliath_best_goliath_mIoU_7673_epoch_194_torchscript.pt2",
+        "0.6b": "sapiens_0.6b_goliath_best_goliath_mIoU_7777_epoch_178_torchscript.pt2",
+        "1b": "sapiens_1b_goliath_best_goliath_mIoU_7994_epoch_151_torchscript.pt2",
+    }
+
+class ModelManager:
+    @staticmethod
+    def load_model(checkpoint_name: str):
+        checkpoint_path = os.path.join(Config.CHECKPOINTS_DIR, Config.CHECKPOINTS[checkpoint_name])
+        model = torch.jit.load(checkpoint_path)
+        model.eval()
+        model.to("cuda")
+        return model
+
+    @staticmethod
+    @torch.inference_mode()
+    def run_model(model, input_tensor, height, width):
+        output = model(input_tensor)
+        output = F.interpolate(output, size=(height, width), mode="bilinear", align_corners=False)
+        _, preds = torch.max(output, 1)
+        return preds
+
+class ImageProcessor:
+    def __init__(self):
+        self.transform_fn = transforms.Compose([
+            transforms.Resize((1024, 768)),
+            transforms.ToTensor(),
+            transforms.Normalize(mean=[123.5/255, 116.5/255, 103.5/255], std=[58.5/255, 57.0/255, 57.5/255]),
+        ])
+
+    @spaces.GPU
+    def process_image(self, image: Image.Image, model_name: str):
+        model = ModelManager.load_model(model_name)
+        input_tensor = self.transform_fn(image).unsqueeze(0).to("cuda")
+        
+        preds = ModelManager.run_model(model, input_tensor, image.height, image.width)
+        mask = preds.squeeze(0).cpu().numpy()
+
+        # Visualize the segmentation
+        blended_image = self.visualize_pred_with_overlay(image, mask)
+
+        # Create downloadable .npy file
+        npy_path = tempfile.mktemp(suffix='.npy')
+        np.save(npy_path, mask)
+
+        return blended_image, npy_path
+
+    @staticmethod
+    def visualize_pred_with_overlay(img, sem_seg, alpha=0.5):
+        img_np = np.array(img.convert("RGB"))
+        sem_seg = np.array(sem_seg)
+
+        num_classes = len(GOLIATH_CLASSES)
+        ids = np.unique(sem_seg)[::-1]
+        legal_indices = ids < num_classes
+        ids = ids[legal_indices]
+        labels = np.array(ids, dtype=np.int64)
+
+        colors = [GOLIATH_PALETTE[label] for label in labels]
+
+        overlay = np.zeros((*sem_seg.shape, 3), dtype=np.uint8)
+
+        for label, color in zip(labels, colors):
+            overlay[sem_seg == label, :] = color
+
+        blended = np.uint8(img_np * (1 - alpha) + overlay * alpha)
+        return Image.fromarray(blended)
+
+class GradioInterface:
+    def __init__(self):
+        self.image_processor = ImageProcessor()
+
+    def create_interface(self):
+        app_styles = """
+        <style>
+            /* Global Styles */
+            body, #root {
+                font-family: Helvetica, Arial, sans-serif;
+                background-color: #1a1a1a;
+                color: #fafafa;
+            }
+            /* Header Styles */
+            .app-header {
+                background: linear-gradient(45deg, #1a1a1a 0%, #333333 100%);
+                padding: 24px;
+                border-radius: 8px;
+                margin-bottom: 24px;
+                text-align: center;
+            }
+            .app-title {
+                font-size: 48px;
+                margin: 0;
+                color: #fafafa;
+            }
+            .app-subtitle {
+                font-size: 24px;
+                margin: 8px 0 16px;
+                color: #fafafa;
+            }
+            .app-description {
+                font-size: 16px;
+                line-height: 1.6;
+                opacity: 0.8;
+                margin-bottom: 24px;
+            }
+            /* Button Styles */
+            .publication-links {
+                display: flex;
+                justify-content: center;
+                flex-wrap: wrap;
+                gap: 8px;
+                margin-bottom: 16px;
+            }
+            .publication-link {
+                display: inline-flex;
+                align-items: center;
+                padding: 8px 16px;
+                background-color: #333;
+                color: #fff !important;
+                text-decoration: none !important;
+                border-radius: 20px;
+                font-size: 14px;
+                transition: background-color 0.3s;
+            }
+            .publication-link:hover {
+                background-color: #555;
+            }
+            .publication-link i {
+                margin-right: 8px;
+            }
+            /* Content Styles */
+            .content-container {
+                background-color: #2a2a2a;
+                border-radius: 8px;
+                padding: 24px;
+                margin-bottom: 24px;
+            }
+            /* Image Styles */
+            /* Updated Image Styles */
+            .image-preview img {
+                max-width: 512px;
+                max-height: 512px;  
+                margin: 0 auto;
+                border-radius: 4px;
+                display: block;
+                object-fit: contain;  
+            }
+
+            /* Control Styles */
+            .control-panel {
+                background-color: #333;
+                padding: 16px;
+                border-radius: 8px;
+                margin-top: 16px;
+            }
+            /* Gradio Component Overrides */
+            .gr-button {
+                background-color: #4a4a4a;
+                color: #fff;
+                border: none;
+                border-radius: 4px;
+                padding: 8px 16px;
+                cursor: pointer;
+                transition: background-color 0.3s;
+            }
+            .gr-button:hover {
+                background-color: #5a5a5a;
+            }
+            .gr-input, .gr-dropdown {
+                background-color: #3a3a3a;
+                color: #fff;
+                border: 1px solid #4a4a4a;
+                border-radius: 4px;
+                padding: 8px;
+            }
+            .gr-form {
+                background-color: transparent;
+            }
+            .gr-panel {
+                border: none;
+                background-color: transparent;
+            }
+            /* Override any conflicting styles from Bulma */
+            .button.is-normal.is-rounded.is-dark {
+                color: #fff !important;
+                text-decoration: none !important;
+            }
+        </style>
+        """
+
+        header_html = f"""
+        <link rel="stylesheet" href="https://cdn.jsdelivr.net/npm/bulma@0.9.3/css/bulma.min.css">
+        <link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.4/css/all.css">
+        {app_styles}
+        <div class="app-header">
+            <h1 class="app-title">Sapiens:Body-Part Segmentation</h1>
+            <h2 class="app-subtitle">ECCV 2024 (Oral)</h2>
+            <p class="app-description">
+                Meta presents Sapiens, foundation models for human tasks pretrained on 300 million human images. 
+                This demo showcases the finetuned body-part segmentation model. <br>
+            </p>
+            <div class="publication-links">
+                <a href="https://arxiv.org/abs/2408.12569" class="publication-link">
+                    <i class="fas fa-file-pdf"></i>arXiv
+                </a>
+                <a href="https://github.com/facebookresearch/sapiens" class="publication-link">
+                    <i class="fab fa-github"></i>Code
+                </a>
+                <a href="https://about.meta.com/realitylabs/codecavatars/sapiens/" class="publication-link">
+                    <i class="fas fa-globe"></i>Meta
+                </a>
+                <a href="https://rawalkhirodkar.github.io/sapiens" class="publication-link">
+                    <i class="fas fa-chart-bar"></i>Results
+                </a>
+            </div>
+            <div class="publication-links">
+                <a href="https://huggingface.co/spaces/facebook/sapiens_pose" class="publication-link">
+                    <i class="fas fa-user"></i>Demo-Pose
+                </a>
+                <a href="https://huggingface.co/spaces/facebook/sapiens_seg" class="publication-link">
+                    <i class="fas fa-puzzle-piece"></i>Demo-Seg
+                </a>
+                <a href="https://huggingface.co/spaces/facebook/sapiens_depth" class="publication-link">
+                    <i class="fas fa-cube"></i>Demo-Depth
+                </a>
+                <a href="https://huggingface.co/spaces/facebook/sapiens_normal" class="publication-link">
+                    <i class="fas fa-vector-square"></i>Demo-Normal
+                </a>
+            </div>
+        </div>
+        """
+
+        js_func = """
+        function refresh() {
+            const url = new URL(window.location);
+            if (url.searchParams.get('__theme') !== 'dark') {
+                url.searchParams.set('__theme', 'dark');
+                window.location.href = url.href;
+            }
+        }
+        """
+
+        def process_image(image, model_name):
+            result, npy_path = self.image_processor.process_image(image, model_name)
+            return result, npy_path
+
+        with gr.Blocks(js=js_func, theme=gr.themes.Default()) as demo:
+            gr.HTML(header_html)
+            with gr.Row(elem_classes="content-container"):
+                with gr.Column():
+                    input_image = gr.Image(label="Input Image", type="pil", format="png", elem_classes="image-preview")
+                    model_name = gr.Dropdown(
+                        label="Model Size",
+                        choices=list(Config.CHECKPOINTS.keys()),
+                        value="1b",
+                    )
+                    example_model = gr.Examples(
+                        inputs=input_image,
+                        examples_per_page=14,
+                        examples=[
+                            os.path.join(Config.ASSETS_DIR, "images", img)
+                            for img in os.listdir(os.path.join(Config.ASSETS_DIR, "images"))
+                        ],
+                    )
+                with gr.Column():
+                    result_image = gr.Image(label="Segmentation Result", type="pil", elem_classes="image-preview")
+                    npy_output = gr.File(label="Segmentation (.npy)")
+                    run_button = gr.Button("Run")
+                    gr.Image(os.path.join(Config.ASSETS_DIR, "palette.jpg"), label="Class Palette", type="filepath", elem_classes="image-preview")
+
+            run_button.click(
+                fn=process_image,
+                inputs=[input_image, model_name],
+                outputs=[result_image, npy_output],
+            )
+            
+        return demo
+
+def main():
+    # Configure CUDA if available
+    if torch.cuda.is_available() and torch.cuda.get_device_properties(0).major >= 8:
+        torch.backends.cuda.matmul.allow_tf32 = True
+        torch.backends.cudnn.allow_tf32 = True
+
+    interface = GradioInterface()
+    demo = interface.create_interface()
+    demo.launch(share=False)
+
+if __name__ == "__main__":
+    main()

assets/checkpoints/sapiens_0.3b_goliath_best_goliath_mIoU_7673_epoch_194_torchscript.pt2

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:735a9a8d63fe8f3f6a4ca3d787de07e69b1f9708ad550e09bb33c9854b7eafbc
+size 1358871599

assets/checkpoints/sapiens_0.6b_goliath_best_goliath_mIoU_7777_epoch_178_torchscript.pt2

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:86aa2cb9d7310ba1cb1971026889f1d10d80ddf655d6028aea060aae94d82082
+size 2685144079

assets/checkpoints/sapiens_1b_goliath_best_goliath_mIoU_7994_epoch_151_torchscript.pt2

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:33bba30f3de8d9cfd44e4eaa4817b1bfdd98c188edfc87fa7cc031ba0f4edc17
+size 4716314057

assets/checkpoints/sapiens_2b_goliath_best_goliath_mIoU_8131_epoch_200_torchscript.pt2

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:afe0970265f2af97f9eeb625036f147730d56820d6891803b13a278160c0f98a
+size 8706620345

classes_and_palettes.py

@@ -0,0 +1,93 @@
+ORIGINAL_GOLIATH_CLASSES = (
+    "Background",
+    "Apparel",
+    "Chair",
+    "Eyeglass_Frame",
+    "Eyeglass_Lenses",
+    "Face_Neck",
+    "Hair",
+    "Headset",
+    "Left_Foot",
+    "Left_Hand",
+    "Left_Lower_Arm",
+    "Left_Lower_Leg",
+    "Left_Shoe",
+    "Left_Sock",
+    "Left_Upper_Arm",
+    "Left_Upper_Leg",
+    "Lower_Clothing",
+    "Lower_Spandex",
+    "Right_Foot",
+    "Right_Hand",
+    "Right_Lower_Arm",
+    "Right_Lower_Leg",
+    "Right_Shoe",
+    "Right_Sock",
+    "Right_Upper_Arm",
+    "Right_Upper_Leg",
+    "Torso",
+    "Upper_Clothing",
+    "Visible_Badge",
+    "Lower_Lip",
+    "Upper_Lip",
+    "Lower_Teeth",
+    "Upper_Teeth",
+    "Tongue",
+)
+
+ORIGINAL_GOLIATH_PALETTE = [
+    [50, 50, 50],
+    [255, 218, 0],
+    [102, 204, 0],
+    [14, 0, 204],
+    [0, 204, 160],
+    [128, 200, 255],
+    [255, 0, 109],
+    [0, 255, 36],
+    [189, 0, 204],
+    [255, 0, 218],
+    [0, 160, 204],
+    [0, 255, 145],
+    [204, 0, 131],
+    [182, 0, 255],
+    [255, 109, 0],
+    [0, 255, 255],
+    [72, 0, 255],
+    [204, 43, 0],
+    [204, 131, 0],
+    [255, 0, 0],
+    [72, 255, 0],
+    [189, 204, 0],
+    [182, 255, 0],
+    [102, 0, 204],
+    [32, 72, 204],
+    [0, 145, 255],
+    [14, 204, 0],
+    [0, 128, 72],
+    [204, 0, 43],
+    [235, 205, 119],
+    [115, 227, 112],
+    [157, 113, 143],
+    [132, 93, 50],
+    [82, 21, 114],
+]
+
+## 6 classes to remove
+REMOVE_CLASSES = (
+    "Eyeglass_Frame",
+    "Eyeglass_Lenses",
+    "Visible_Badge",
+    "Chair",
+    "Lower_Spandex",
+    "Headset",
+)
+
+## 34 - 6 = 28 classes left
+GOLIATH_CLASSES = tuple(
+    [x for x in ORIGINAL_GOLIATH_CLASSES if x not in REMOVE_CLASSES]
+)
+GOLIATH_PALETTE = [
+    ORIGINAL_GOLIATH_PALETTE[idx]
+    for idx in range(len(ORIGINAL_GOLIATH_CLASSES))
+    if ORIGINAL_GOLIATH_CLASSES[idx] not in REMOVE_CLASSES
+]

requirements.txt

@@ -0,0 +1,8 @@
+gradio
+numpy
+torch
+torchvision
+matplotlib
+pillow
+spaces
+opencv-python