DL3DV-ALL-960P: 960P Frames with Camera Poses for 3D Vision

DL3DV-Dataset provides all 960P frames with associated camera poses from the DL3DV-10K Dataset. The dataset authors are actively reviewing the content to avoid sensitive information. The repository was last updated on September 2, 2024.

Use Cases

• Train neural radiance field (NeRF) models based on multiview frames and camera poses.
• Develop 3D scene reconstruction algorithms based on posed image sequences.
• Benchmark novel view synthesis methods based on high-resolution (960P) video frames.
• Fine-tune depth estimation models based on synchronized camera pose data.

Strengths

• Contains high-resolution 960P frames, as stated in the title and description.
• Includes camera pose data for each frame, a key feature for 3D vision tasks.
• Authors are conducting a review for sensitive content, suggesting attention to data ethics.

Limitations

• Description metadata is limited; actual data quality requires manual inspection after download.
• Column-level documentation is absent; field semantics must be inferred after download.
• Row count, file formats, and license are unknown, which may limit suitability assessment.

Provenance

Source

DL3DV

Collection Method

Likely collected or synthesized for 3D vision research.

Time Range

null

Freshness

Last updated 2024-09-02 19:11:31; freshness should be verified.

Geography

null