README.md

---
dataset_info:
  - config_name: default
    data_files:
      - split: train
        path: train/*
      - split: test_seen
        path: test_seen/*
      - split: test_unseen
        path: test_unseen/*
tags:
  - video-generation
  - multimedia
  - benchmark
  - evaluation
  - camera-control
license: mit
language:
  - en
  - zh
task_categories:
  - video-to-video
---

# FollowBench

## Dataset Description

**FollowBench** is a benchmark for evaluating video generation models on viewpoint transformation and subject following tasks. It features diverse scenarios and appearances to test novel view synthesis and geometric spatial understanding capabilities.

## 📊 Data Splits
| Split | Samples | Description |
|-------|-----------|-------------|
| `train` | 8644 | Training set for model optimization and development |
| `test_seen` | 1235 | In-distribution test set for evaluating performance on familiar scenarios |
| `test_unseen` | 1138 | Out-of-distribution test set for assessing generalization to novel scenarios |