Sign in to view source links and access this dataset
Description
Vista4D is an evaluation dataset for video reshooting with 4D point clouds, associated with a CVPR 2026 Highlight paper. The dataset was created by a consortium including Eyeline Labs, Netflix, Columbia University, UCLA, Stony Brook University, and the University of Oxford. It was last updated on HuggingFace on April 24, 2026.
Use Cases
Benchmarking video reshooting algorithms based on the described 4D point cloud data.
Evaluating novel view synthesis for dynamic scenes based on the video reshooting task.
Training or testing 4D reconstruction models based on the provided evaluation data.
Strengths
Dataset is associated with a peer-reviewed CVPR 2026 Highlight paper.
Created by a multi-institutional team including industry (Netflix, Eyeline Labs) and academic (Columbia, UCLA, Oxford) partners.
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 sample data are unknown, which may limit suitability assessment.
Provenance
Source
Eyeline-Labs on HuggingFace
Collection Method
Likely created for evaluating the Vista4D video reshooting method.
Time Range
null
Freshness
Last updated 2026-04-24 01:56:45; freshness should be verified.
Geography
null
License is unknown; users must verify licensing terms before use.