Data Engine Feature

.gitmodules

@@ -1,3 +1,6 @@
 [submodule "third_party/Depth-Anything-ONNX"]
 	path = third_party/Depth-Anything-ONNX
 	url = https://github.com/fabio-sim/Depth-Anything-ONNX.git
+[submodule "third_party/sam2"]
+	path = third_party/sam2
+	url = https://github.com/facebookresearch/sam2.git

.pre-commit-config.yaml

@@ -5,7 +5,6 @@ repos:
       - id: trailing-whitespace
       - id: end-of-file-fixer
       - id: check-yaml
-      - id: check-added-large-files
       - id: check-merge-conflict
 
   - repo: https://github.com/psf/black

data_engine/Readme.md

@@ -0,0 +1,3 @@
+# Data Engine
+
+For now we have a POC of the project. Data engine is in progress of development.

data_engine/poc/.dockerignore

@@ -0,0 +1,118 @@
+# Docker ignore file for SAM2 Data Engine
+
+# Python
+__pycache__/
+*.py[cod]
+*$py.class
+*.so
+.Python
+build/
+develop-eggs/
+dist/
+downloads/
+eggs/
+.eggs/
+lib/
+lib64/
+parts/
+sdist/
+var/
+wheels/
+*.egg-info/
+.installed.cfg
+*.egg
+MANIFEST
+
+# Virtual environments
+venv/
+env/
+ENV/
+env.bak/
+venv.bak/
+
+# Cache directories
+cache/
+.cache/
+*.cache
+
+# IDE and editor files
+.vscode/
+.idea/
+*.swp
+*.swo
+*~
+.DS_Store
+Thumbs.db
+
+# Logs
+*.log
+logs/
+
+# Test files
+test_output/
+test_config/
+.pytest_cache/
+.coverage
+htmlcov/
+
+# Model files (large)
+*.pt
+*.pth
+*.onnx
+*.trt
+models/
+
+# Data directories
+data/
+datasets/
+videos/
+exports/
+projects/
+
+# Docker files (don't copy into container)
+Dockerfile*
+docker-compose*.yml
+.dockerignore
+
+# Git
+.git/
+.gitignore
+
+# Documentation
+docs/
+*.md
+!README.md
+
+# Temporary files
+tmp/
+temp/
+.tmp/
+
+# OS specific
+.DS_Store
+Thumbs.db
+desktop.ini
+
+# Jupyter
+.ipynb_checkpoints/
+*.ipynb
+
+# Environment files
+.env
+.env.local
+.env.*.local
+
+# Large media files
+*.mp4
+*.avi
+*.mov
+*.mkv
+*.wmv
+*.webm
+*.flv
+
+# Archives
+*.zip
+*.tar.gz
+*.rar
+*.7z

data_engine/poc/Dockerfile

@@ -0,0 +1,89 @@
+# SAM2 Data Engine Dockerfile
+FROM pytorch/pytorch:2.7.1-cuda12.6-cudnn9-runtime
+
+# Install system dependencies for GUI applications and git
+RUN apt-get update && apt-get install -y \
+    libgl1-mesa-glx \
+    libglib2.0-0 \
+    libxrender1 \
+    libxrandr2 \
+    libxss1 \
+    libxcursor1 \
+    libxcomposite1 \
+    libasound2 \
+    libxi6 \
+    libxtst6 \
+    libqt5gui5 \
+    libqt5core5a \
+    libqt5widgets5 \
+    qt5-gtk-platformtheme \
+    libxcb1 \
+    libxcb-xkb1 \
+    libxkbcommon-x11-0 \
+    libxcb-icccm4 \
+    libxcb-image0 \
+    libxcb-keysyms1 \
+    libxcb-randr0 \
+    libxcb-render-util0 \
+    libxcb-render0 \
+    libxcb-shape0 \
+    libxcb-sync1 \
+    libxcb-xfixes0 \
+    libxcb-xinerama0 \
+    libxcb-xinput0 \
+    libxcb-cursor0 \
+    x11-apps \
+    git \
+    wget \
+    && rm -rf /var/lib/apt/lists/*
+
+# Set working directory
+WORKDIR /app
+
+# Clone SAM2 repository
+RUN git clone https://github.com/facebookresearch/segment-anything-2.git /app/third_party/sam2
+
+# Copy requirements first for better caching
+COPY requirements.txt .
+
+# Install Python dependencies including SAM2
+RUN pip install --no-cache-dir -r requirements.txt
+RUN cd /app/third_party/sam2 && pip install -e .
+
+# Create models directory and config directory
+RUN mkdir -p /app/models /app/.config /app/sam2_models
+
+# Set environment variables for Ultralytics
+ENV YOLO_CONFIG_DIR=/app/.config
+ENV YOLO_CONFIG_DIR=/app/.config
+
+# Download SAM2 models during build time
+RUN python3 -c "\
+import sys; \
+sys.path.append('/app/third_party/sam2'); \
+import os; \
+import urllib.request; \
+os.makedirs('/app/sam2_models', exist_ok=True); \
+checkpoint_urls = { \
+    'sam2_hiera_tiny.pt': 'https://dl.fbaipublicfiles.com/segment_anything_2/072824/sam2_hiera_tiny.pt', \
+    'sam2_hiera_small.pt': 'https://dl.fbaipublicfiles.com/segment_anything_2/072824/sam2_hiera_small.pt', \
+    'sam2_hiera_base_plus.pt': 'https://dl.fbaipublicfiles.com/segment_anything_2/072824/sam2_hiera_base_plus.pt', \
+    'sam2_hiera_large.pt': 'https://dl.fbaipublicfiles.com/segment_anything_2/072824/sam2_hiera_large.pt' \
+}; \
+[urllib.request.urlretrieve(url, f'/app/sam2_models/{filename}') for filename, url in checkpoint_urls.items() if not os.path.exists(f'/app/sam2_models/{filename}')]; \
+print('All SAM2 models downloaded')"
+
+# Download YOLOv8 models during build time (keeping YOLO for dataset export)
+RUN python -c "from ultralytics import YOLO; model = YOLO('yolov8n.pt'); print('YOLOv8 nano model downloaded')"
+RUN python -c "from ultralytics import YOLO; model = YOLO('yolov8s.pt'); print('YOLOv8 small model downloaded')"
+RUN python -c "from ultralytics import YOLO; model = YOLO('yolov8m.pt'); print('YOLOv8 medium model downloaded')"
+
+# Copy the current directory contents
+COPY . .
+
+# Set environment variables for GUI
+ENV QT_X11_NO_MITSHM=1
+ENV DISPLAY=:0
+
+# Run the main GUI application
+CMD ["python", "main.py"]

data_engine/poc/README.md

@@ -0,0 +1,30 @@
+# SAM2 Data Engine - YOLO Dataset Generator
+
+An automatic data engine GUI application that uses visual prompts to interact with SAM2 (Segment Anything Model 2) for generating YOLO training datasets with automatic video segmentation.
+
+
+
+This implementation is a POC of the actual data engine that we want to build. It should serve as reference as the type of features we want to have.
+
+
+### How to run this code?
+
+It has a dockerfile to run the application.
+
+First give access for the docker access the xserver with this command:
+
+```bash
+xhost +
+```
+
+
+Then run docker container with compose. Inside the directory run the following command:
+
+```bash
+docker compose up
+```
+
+You should see the following applicatation running.
+
+
+![](poc_screenshot.png)