Camera Motion Annotation Dataset
Access 100,000 professionally curated video clips with human-generated, frame-accurate camera motion annotations. The collection provides approximately 417 hours of annotated 4K video designed for training and evaluating models that recognize, classify, and understand camera movement patterns across diverse real-world scenes.
Each clip includes quality-controlled temporal annotations identifying camera motion type with precise start and end boundaries. Structured metadata includes motion labels, frame rate, resolution, content category, and natural-language scene descriptions, providing supervised data for video understanding, temporal segmentation, scene analysis, retrieval, and motion-aware generation. Motion-labelled video of this kind is used alongside our broader computer vision datasets.
Dataset Preview
Representative video examples showing source scenes and camera movements together with corresponding frame-accurate temporal annotations.
The examples illustrate how human-generated motion labels align with the source video, including the temporal boundaries of individual camera movements.
Key Highlights
- • Frame-accurate camera motion start and end boundaries
- • Human-generated, quality-controlled camera motion labels
- • Clean source video without graphics, captions, or overlays
- • Structured CSV annotations aligned to individual video assets
- • Rights-cleared professionally produced video content
Metadata Fields
Technical Specifications
Dataset Type
Human-Annotated Camera Motion Video Dataset
Clip Count
100,000 annotated clips
Total Duration
Approximately 417 hours
Average Clip Length
Approximately 15 seconds
Video Resolution
4K
Frame Rate
24 fps
Annotation Type
Human-generated camera motion labels
Annotation Precision
Frame-accurate temporal start and end boundaries
Annotation Format
CSV
Annotation Alignment
Individual CSV annotation data aligned to corresponding video assets
Scene Metadata
Content category and natural-language scene descriptions
Source Quality
Professionally produced video
Source Presentation
Clean video without graphics, captions, or overlays
Audio
None
Licensing, Documentation, and Delivery
Wavebreak Media can provide applicable licensing, provenance, and model or property release information for the rights-cleared camera motion video and annotations, with dataset delivery organized around agreed model-development, benchmarking, and delivery requirements. See our Dataset Licensing & Compliance and Dataset Delivery & Security pages for details on rights documentation, packaging, and secure transfer.
Need the Full Dataset?
Request access to the Camera Motion Annotation Dataset or discuss annotation structure, motion labels, metadata, licensing, benchmarking, and technical delivery requirements with Wavebreak Media.
Related Datasets
Synchronized Multi-View Video Dataset
Access synchronized 4K multi-camera video captured from three viewpoints for multi-view synthesis, camera-aware generation, shot planning, cinematic understanding, and video editing AI.
View Dataset CONTROLLED HUMAN MOTION CAPTUREHuman Motion Dataset (DV01)
Access about 6,000 purpose-built human motion videos with structured metadata for action recognition, gesture understanding, behaviour modelling, video understanding, and multimodal AI.
View Dataset PEDESTRIAN & CITY INTERACTION CAPTUREUrban City Life & Mobility Video Dataset
Access 1,500 purpose-built 4K urban videos capturing pedestrians, commuting, city navigation, and social interactions for mobility analysis, activity recognition, and computer vision.
View Dataset