Add video perception datasets

README.md

@@ -9,67 +9,67 @@ language:
 tags:
 - vision
 - agents
-- gui
-- web
-- spatial-reasoning
-- game-understanding
-- robotics
-- manipulation
+- video-understanding
+- egocentric
 size_categories:
 - 1K<n<10K
 configs:
-- config_name: 3dsrbench
-  data_files:
-  - split: train
-    path: 3dsrbench/train-*
 - config_name: android-in-the-wild
   data_files:
   - split: train
     path: android-in-the-wild/train-*
-- config_name: but-they-are-cats-tutorial
+- config_name: mind2web
   data_files:
   - split: train
-    path: but-they-are-cats-tutorial/train-*
-- config_name: embspatial-bench
+    path: mind2web/train-*
+- config_name: monday
   data_files:
   - split: train
-    path: embspatial-bench/train-*
-- config_name: erqa
+    path: monday/train-*
+- config_name: screenspot
   data_files:
   - split: train
-    path: erqa/train-*
-- config_name: gameqa
+    path: screenspot/train-*
+- config_name: screenspot-pro
   data_files:
   - split: train
-    path: gameqa/train-*
-- config_name: manipbench
+    path: screenspot-pro/train-*
+- config_name: embspatial-bench
   data_files:
   - split: train
-    path: manipbench/train-*
-- config_name: marioqa
+    path: embspatial-bench/train-*
+- config_name: space-10
   data_files:
   - split: train
-    path: marioqa/train-*
-- config_name: mind2web
+    path: space-10/train-*
+- config_name: 3dsrbench
   data_files:
   - split: train
-    path: mind2web/train-*
-- config_name: monday
+    path: 3dsrbench/train-*
+- config_name: spatialmm
   data_files:
   - split: train
-    path: monday/train-*
+    path: spatialmm/train-*
 - config_name: omnispatial
   data_files:
   - split: train
     path: omnispatial/train-*
-- config_name: openeqa
+- config_name: gameqa
   data_files:
   - split: train
-    path: openeqa/train-*
-- config_name: perception-test
+    path: gameqa/train-*
+- config_name: videogamebunny
   data_files:
   - split: train
-    path: perception-test/train-*
+    path: videogamebunny/train-*
+- config_name: but-they-are-cats-tutorial
+  data_files:
+  - split: train
+    path: but-they-are-cats-tutorial/train-*
+- config_name: marioqa
+  data_files:
+  - split: train
+    path: marioqa/train-*
 - config_name: robo2vlm
   data_files:
   - split: train
@@ -78,405 +78,51 @@ configs:
   data_files:
   - split: train
     path: roborefit/train-*
-- config_name: screenspot
+- config_name: erqa
   data_files:
   - split: train
-    path: screenspot/train-*
-- config_name: screenspot-pro
+    path: erqa/train-*
+- config_name: manipbench
   data_files:
   - split: train
-    path: screenspot-pro/train-*
+    path: manipbench/train-*
 - config_name: sharerobot
   data_files:
   - split: train
     path: sharerobot/train-*
-- config_name: space-10
+- config_name: perception-test
   data_files:
   - split: train
-    path: space-10/train-*
-- config_name: spatialmm
+    path: perception-test/train-*
+- config_name: vsi-bench
   data_files:
   - split: train
-    path: spatialmm/train-*
-- config_name: videogamebunny
+    path: vsi-bench/train-*
+- config_name: openeqa
   data_files:
   - split: train
-    path: videogamebunny/train-*
-- config_name: vsi-bench
+    path: openeqa/train-*
+- config_name: videgothink
   data_files:
   - split: train
-    path: vsi-bench/train-*
-dataset_info:
-- config_name: openeqa
-  features:
-  - name: Video
-    dtype: video
-  - name: Question
-    dtype: string
-  - name: Options
-    list: string
-  - name: Answer
-    dtype: int64
-  - name: Category
-    dtype: string
-  splits:
-  - name: train
-    num_bytes: 42234040347.0
-    num_examples: 1000
-  download_size: 539548
-  dataset_size: 42234040347.0
-- config_name: perception-test
-  features:
-  - name: Video
-    dtype: video
-  - name: Question
-    dtype: string
-  - name: Options
-    list: string
-  - name: Answer
-    dtype: int64
-  splits:
-  - name: train
-    num_bytes: 6522333449.0
-    num_examples: 997
-  download_size: 216600
-  dataset_size: 6522333449.0
-- config_name: sharerobot
-  features:
-  - name: Images_Base64
-    list: string
-  - name: Image_Count
-    dtype: int32
-  - name: Question
-    dtype: string
-  - name: Options
-    list: string
-  - name: Answer
-    dtype: int32
-  - name: Category
-    dtype: string
-  splits:
-  - name: train
-    num_bytes: 564326225
-    num_examples: 1000
-  download_size: 563906378
-  dataset_size: 564326225
-- config_name: vsi-bench
-  features:
-  - name: Video
-    dtype: video
-  - name: Question
-    dtype: string
-  - name: Options
-    list: string
-  - name: Answer
-    dtype: int64
-  - name: Category
-    dtype: string
-  splits:
-  - name: train
-    num_bytes: 15051493138.0
-    num_examples: 1000
-  download_size: 434406
-  dataset_size: 15051493138.0
+    path: videgothink/train-*
+- config_name: egoplan-bench
+  data_files:
+  - split: train
+    path: egoplan-bench/train-*
+
 ---
 
 # AgentVQA: A Multi-Domain Visual Question Answering Dataset
 
-AgentVQA is a comprehensive dataset for training and evaluating visual agents across multiple domains, including GUI interaction, spatial reasoning, game understanding, and robot manipulation. The dataset contains multiple-choice questions based on screenshots, images, and videos.
-
-## Dataset Structure
-
-The dataset is organized into four main domains:
-
-### 1. Web Agents Domain (GUI Interaction)
-
-- **android-in-the-wild**: Android app interactions (1000 samples)
-- **mind2web**: Web navigation tasks (1000 samples, stratified by category)
-- **monday**: Mobile OS interactions (1000 samples, stratified by OS)
-- **screenspot**: GUI element grounding (1000 samples)
-- **screenspot-pro**: Advanced GUI grounding (1000 samples)
-
-### 2. Spatial Reasoning Domain
-
-- **embspatial-bench**: Object-object and object-scene spatial relationships (1000 samples)
-- **space-10**: Multi-image spatial reasoning with entity presence tasks (1000 samples)
-- **3dsrbench**: 3D spatial reasoning benchmark
-- **spatialmm**: Spatial reasoning in multi-modal contexts
-- **omnispatial**: Comprehensive spatial reasoning across multiple task types
-
-### 3. Game Understanding Domain
-
-- **gameqa**: 3D spatial perception and reasoning in games (1000 samples)
-- **videogamebunny**: Video game understanding across multiple categories
-- **but-they-are-cats-tutorial**: Game tutorial comprehension tasks
-- **marioqa**: Video-based event understanding in Super Mario gameplay
-
-### 4. Robot Manipulation Domain
-
-- **robo2vlm**: Robot task success state evaluation (1000 samples)
-- **roborefit**: Robot grasping and manipulation tasks (1000 samples)
-- **erqa**: Embodied reasoning for robot question answering (400 samples)
-- **manipbench**: Robot manipulation trajectory planning (1000 samples)
-- **sharerobot**: Multi-frame robot task understanding (1000 samples)
-
 ## Loading the Dataset
 
-### Load a specific dataset:
-
 ```python
 from datasets import load_dataset
 
-# Load a web agents dataset
-ds = load_dataset("advaitgupta/AgentVQA", "android-in-the-wild", split="train")
-
-# Load a spatial reasoning dataset
-ds = load_dataset("advaitgupta/AgentVQA", "embspatial-bench", split="train")
-
-# Load a game understanding dataset
-ds = load_dataset("advaitgupta/AgentVQA", "gameqa", split="train")
-
-# Load a robot manipulation dataset
-ds = load_dataset("advaitgupta/AgentVQA", "robo2vlm", split="train")
+ds = load_dataset("advaitgupta/AgentVQA", "perception-test", split="train")
+sample = ds[0]
 ```
 
-### Helper Functions:
-
-```python
-from datasets import load_dataset
-
-# Available domains and their datasets
-DOMAINS = {
-    "web_agents": [
-        "android-in-the-wild",
-        "mind2web",
-        "monday",
-        "screenspot",
-        "screenspot-pro"
-    ],
-    "spatial_reasoning": [
-        "embspatial-bench",
-        "space-10",
-        "3dsrbench",
-        "spatialmm",
-        "omnispatial"
-    ],
-    "game_understanding": [
-        "gameqa",
-        "videogamebunny",
-        "but-they-are-cats-tutorial",
-        "marioqa"
-    ],
-    "robot_manipulation": [
-        "robo2vlm",
-        "roborefit",
-        "erqa",
-        "manipbench",
-        "sharerobot"
-    ]
-}
-
-def load_domain_dataset(domain, dataset_name):
-    """
-    Load a specific dataset from a domain.
-    
-    Args:
-        domain (str): Domain name (e.g., "web_agents", "robot_manipulation")
-        dataset_name (str): Dataset name (e.g., "android-in-the-wild", "robo2vlm")
-    
-    Returns:
-        Dataset object
-    """
-    return load_dataset("advaitgupta/AgentVQA", dataset_name, split="train")
-
-def load_all_domain_datasets(domain):
-    """
-    Load all datasets from a specific domain.
-    
-    Args:
-        domain (str): Domain name (e.g., "web_agents", "robot_manipulation")
-    
-    Returns:
-        dict: Dictionary mapping dataset names to Dataset objects
-    """
-    datasets = {}
-    for dataset_name in DOMAINS[domain]:
-        datasets[dataset_name] = load_dataset("advaitgupta/AgentVQA", dataset_name, split="train")
-    return datasets
-
-def load_all_datasets():
-    """
-    Load all datasets from all domains.
-    
-    Returns:
-        dict: Dictionary mapping dataset names to Dataset objects
-    """
-    all_datasets = {}
-    for domain in DOMAINS:
-        all_datasets.update(load_all_domain_datasets(domain))
-    return all_datasets
-
-# Example usage:
-# Load specific dataset
-ds = load_domain_dataset("robot_manipulation", "robo2vlm")
-
-# Load all robot manipulation datasets
-robot_datasets = load_all_domain_datasets("robot_manipulation")
-print(robot_datasets.keys())
-
-# Load everything
-all_data = load_all_datasets()
-print(f"Total datasets loaded: {len(all_data)}")
-```
-
-## Dataset Fields
-
-### Web Agents Domain
-
-#### android-in-the-wild
-- `image`: Screenshot of the app
-- `step_id`: Step number in the episode
-- `action_history`: Previous actions taken
-- `question`: The question about next action
-- `options`: List of 4 possible actions
-- `correct_answer`: Index of correct option (0-3)
-- `episode_goal`: Overall goal of the episode
-- `source_dataset`: Original source dataset name
-
-#### mind2web
-- `image`: Screenshot of the webpage
-- `question`: The question about next action
-- `options`: List of possible actions
-- `answer`: Index of correct option
-- `task`: Description of the task
-- `action_history`: Previous actions taken
-- `category`: Task category
-- `website`: Website name
-- `domain`: Domain category
-
-#### monday
-- `image`: Screenshot of the mobile interface
-- `goal`: Goal description
-- `question`: The question about next action
-- `options`: List of possible actions
-- `answer`: Index of correct option
-- `action_history`: Previous actions taken
-- `os`: Operating system (Android/iOS)
-
-#### screenspot & screenspot-pro
-- `image`: Screenshot
-- `question`: Instruction to complete
-- `options`: List of tap/click actions
-- `answer`: Index of correct option
-
-### Spatial Reasoning Domain
-
-#### embspatial-bench
-- `Image`: Scene image
-- `Question`: Spatial relationship question
-- `Options`: List of 4 possible answers
-- `Answer`: Index of correct option (0-3)
-- `Category`: High-level category (e.g., "Object-Object Spatial Relationship")
-
-#### space-10
-- `Images`: List of 8 scene images (multi-image reasoning)
-- `Question`: Entity presence question
-- `Options`: List of 4 descriptions
-- `Answer`: Index of correct option (0-3)
-- `Category`: Task category (e.g., "Entity Presence")
-
-#### 3dsrbench
-- `Image`: 3D scene image
-- `Question`: Spatial reasoning question
-- `Options`: List of 4 possible answers
-- `Answer`: Index of correct option (0-3)
-- `Category`: Task category
-
-#### spatialmm
-- `Image`: Scene image
-- `Question`: Spatial reasoning question
-- `Options`: List of 4 possible answers
-- `Answer`: Index of correct option
-
-#### omnispatial
-- `Image`: Scene image
-- `Question`: Spatial reasoning question
-- `Options`: List of possible answers
-- `Answer`: Index of correct option
-- `Task_Type`: Main task type
-- `Sub_Task_Type`: Specific sub-task category
-
-### Game Understanding Domain
-
-#### gameqa
-- `Image`: Game screenshot
-- `Question`: Game state reasoning question
-- `Options`: List of 4 possible answers
-- `Answer`: Index of correct option (0-3)
-- `Game_Type`: Type of game (e.g., "3D Spatial Perception & Reasoning")
-- `Game_Name`: Specific game name
-
-#### videogamebunny
-- `Image`: Game screenshot
-- `Question`: Game understanding question
-- `Options`: List of 4 possible answers
-- `Answer`: Index of correct option (0-3)
-- `Category`: Question category
-
-#### but-they-are-cats-tutorial
-- `Image`: Game tutorial screenshot
-- `Question`: Game mechanics question
-- `Options`: List of 4 possible answers
-- `Answer`: Index of correct option (0-3)
-
-#### marioqa
-- `Video`: Super Mario gameplay video clip
-- `Question`: Event-based question about gameplay
-- `Options`: List of 4 possible answers
-- `Answer`: Correct option letter (A, B, C, or D)
-- `Question_Type`: Type of question (e.g., "event-centric")
-- `Event`: Event type being queried (e.g., "Appear", "Kill")
-
-### Robot Manipulation Domain
-
-#### robo2vlm
-- `Image`: Robot scene image
-- `Question`: Task success evaluation question
-- `Options`: List of 4 possible answers
-- `Answer`: Index of correct option (0-3)
-- `Category`: Question category (e.g., "task_success_state")
-
-#### roborefit
-- `Image`: Robot manipulation scene
-- `Question`: Grasping instruction
-- `Options`: List of 4 coordinate pairs (picking points)
-- `Answer`: Index of correct option (0-3)
-
-#### erqa
-- `Images`: List of robot scene image(s) with trajectory overlay (always a list, contains 1+ images)
-- `Question`: Embodied reasoning question
-- `Options`: List of 4 possible answers
-- `Answer`: Index of correct option (0-3)
-- `Category`: Question type (e.g., "Trajectory Reasoning", "Action Reasoning")
-
-#### manipbench
-- `Image`: Robot manipulation scene with grid overlay
-- `Question`: Trajectory planning question
-- `Options`: List of 4 picking point and trajectory combinations
-- `Answer`: Index of correct option (0-3)
-- `Category`: Task category (e.g., "droid_arti", "droid_pick_place")
-
-#### sharerobot
-- `Images`: List of robot trajectory frames (variable number of images per sample)
-- `Question`: Task understanding question
-- `Options`: List of 4 possible task descriptions
-- `Answer`: Index of correct option (0-3)
-- `Category`: Task type (e.g., "Past_Description_Task")
-
-## Citation
-
-If you use this dataset, please cite the original papers for each component dataset.
-
 ## License
-
 MIT License