O-37: Light Sequence Data Generator

Generates synthetic pattern recognition tasks where lights in a horizontal row must be turned on/off according to specified logical rules. Tasks test understanding of position, patterns, ranges, and relative locations.

Each sample pairs a task (first frame + prompt describing what needs to happen) with its ground truth solution (final frame showing the result + video demonstrating how to achieve it). This structure enables both model evaluation and training.

---

📌 Basic Information

Property	Value
Task ID	O-37
Task	Light Sequence
Category	Abstraction
Resolution	1024×1024 px
FPS	16 fps
Duration	varies
Output	PNG images + MP4 video

---

🚀 Usage

Installation

# Clone the repository
git clone https://github.com/VBVR-DataFactory/O-37_light_sequence_data-generator.git
cd O-37_light_sequence_data-generator

# Install dependencies
pip install -r requirements.txt

Generate Data

# Generate 100 samples
python examples/generate.py --num-samples 100

# Generate with specific seed
python examples/generate.py --num-samples 100 --seed 42

# Generate without videos
python examples/generate.py --num-samples 100 --no-videos

# Custom output directory
python examples/generate.py --num-samples 100 --output data/my_output

Command-Line Options

Argument	Type	Description	Default
--num-samples	int	Number of samples to generate	100
--seed	int	Random seed for reproducibility	Random
--output	str	Output directory	data
--no-videos	flag	Skip video generation	False

---

📖 Task Example

Prompt

The scene shows 8 circular lights in a horizontal row on a white background. Lights on are green with glow; lights off are gray. Initially, some lights are on and some are off.

Your task: Modify the light states so that all lights in the right half of the row (the 4 lights from the right side, counting from left to right) are on (green with glow), and all lights in the left half are off (gray). Turn lights on/off as needed. Lights change from gray to green (with glow) when turned on, and from green to gray (glow disappears) when turned off. Lights stay in fixed positions; only their states change.

Visual

Initial Frame Random initial light states	Animation Lights turning on/off to match pattern	Final Frame Target pattern achieved

---

📖 Task Description

Objective

Modify the states of lights in a row to match a specified pattern based on logical rules such as position, parity (odd/even), spatial ranges, or continuous sequences.

Task Setup

• Number of Lights: 4, 6, 8, or 10 lights in a horizontal row
• Light States: On (colored with glow effect) or Off (gray)
• Color Variety: Multiple colors available (gold/yellow, pink, blue, green, orange, purple, cyan, red, amber)
• Six Task Types:
  • Type 1: Single specific position (e.g., "3rd light")
  • Type 2: Multiple specific positions (e.g., "2nd and 5th lights")
  • Type 3: Mathematical pattern (e.g., "all odd positions")
  • Type 4: Spatial range (e.g., "left half")
  • Type 5: Continuous sequence (e.g., "from 2nd to 5th")
  • Type 6: Relative position (e.g., "leftmost 3 lights")

Key Features

• Position reasoning: Tests understanding of ordinal positions (1st, 2nd, 3rd, etc.)
• Pattern recognition: Identifies odd/even patterns and spatial relationships
• Range understanding: Interprets spatial ranges (halves, continuous sequences)
• Relative localization: Processes relative descriptors (leftmost, rightmost)
• State transitions: Visualizes lights changing between on/off states with glow effects
• Multiple colors: Different colors per task enhance visual variety
• Fixed positions: Lights don't move; only states change

---

📦 Data Format

data/questions/light_sequence_task/light_sequence_00000000/
├── first_frame.png      # Initial state (random light pattern)
├── final_frame.png      # Final state (target pattern)
├── prompt.txt           # Task instructions with pattern specification
├── ground_truth.mp4     # Solution video (16 fps)
└── question_metadata.json # Task metadata

File specifications: Images are 1024×1024 PNG. Videos are MP4 at 16 fps, duration varies by number of state changes.

---

🏷️ Tags

pattern-recognition logical-reasoning position-understanding sequence-patterns state-transitions spatial-ranges ordinal-reasoning

---