Kubric: A scalable dataset generator

By Klaus Greff, Francois Belletti, Lucas Beyer, Carl Doersch, Yilun Du, Daniel Duckworth, David J. Fleet, Dan Gnanapragasam, Florian Golemo, Charles Herrmann and 24 more. Computer Vision and Pattern Recognition, 2022

467 citations recorded by the source corpus.

Data is the driving force of machine learning, with the amount and quality of training data often being more important for the performance of a system than architecture and training details. But collecting, processing and annotating real data at scale is difficult, expensive, and frequently raises additional privacy, fairness and legal concerns. Synthetic data is a powerful tool with the potential to address these shortcomings: 1) it is cheap 2) supports rich ground-truth annotations 3) offers full control over data and 4) can circumvent or mitigate problems regarding bias, privacy and licensing. Unfortunately, software tools for effective data generation are less mature than those for architecture design and training, which leads to fragmented generation efforts. To address these problems we introduce Kubric, an open-source Python framework that interfaces with PyBullet and Blender to generate photo-realistic scenes, with rich annotations, and seamlessly scales to large jobs distributed over thousands of machines, and generating TBs of data. We demonstrate the effectiveness of Kubric by presenting a series of 13 different generated datasets for tasks ranging from studying 3D NeRF models to optical flow estimation. We release Kubric, the used assets, all of the generation code, as well as the rendered datasets for reuse and modification.

Subfields: Computer Vision and Pattern Recognition, Environmental Engineering

View Paper: https://www.semanticscholar.org/paper/7613b1df07b909a91e608d75babc834df40cf85a

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computer Vision and Pattern Recognition

In Environmental Engineering

Citations in the corpus

VGGT: Visual Geometry Grounded Transformer (2025, unknown): “We ﬁnetune the entire modiﬁed tracker on Kubric [29].”
TAPIR: Tracking Any Point with per-frame Initialization and temporal Refinement (2023, background): “One subtle difference between the Kubric MOViE [17] dataset and the real world is the lack of panning: although the Kubric camera moves, it is always set to “look at” a single point at the center of the workspace.”
Gen3C: 3D-Informed World-Consistent Video Generation with Precise Camera Control (2025, unknown): “We select three real-world videos datasets: RE10K [74], DL3DV [29], Waymo Open Dataset (WOD) [49], and a synthetic dataset Kubric4D [14, 53].”
Object Scene Representation Transformer (2022, background): “It is a highly demanding dataset due to its use of photo-realistic ray tracing [8], complex arrangements of tightly-packed objects of varying size, challenging backgrounds, and nontrivial camera poses including almost horizontal views of the scene.”
ProcTHOR: Large-Scale Embodied AI Using Procedural Generation (2022, background): “In the broader field of computer vision, the dataset may be used to study object detection [64]; NeRFs [80, 110, 43, 71]; segmentation, depth, and optimal flow estimation [35, 43]; generative modeling [59, 62, 61]; occlusion reasoning [34]; and pose estimation [19], among others.”
CoTracker3: Simpler and Better Point Tracking by Pseudo-Labeling Real Videos (2025, unknown): “All the methods mentioned above follow this paradigm, and most are trained solely on synthetic Kubric [11].”
TAP-Vid: A Benchmark for Tracking Any Point in a Video (2022, unknown): “Kubric-VFS-Like [18]: A baseline for point tracking trained on Kubric using contrastive learning, inspired by VFS [81].”
PointOdyssey: A Large-Scale Synthetic Dataset for Long-Term Point Tracking (2023, methodology): “MPI Sintel [13] Flyingthings++ [40, 25] Kubric [24] TAP-Vid-Kinetics [16] TAP-Vid-DAVIS [16] Depth & normals Table 1: Comparison of point tracking datasets.”
Mind's Eye: Grounded Language Model Reasoning through Simulation (2022, background): “World models should be capable of planning (Li et al., 2022a), predicting (Amos et al., 2018 2022)), while our focus is grounded reasoning in the form of natural language.”
Perception Test: A Diagnostic Benchmark for Multimodal Video Models (2023, methodology): “Point tracking: We report baseline results using a TAP-Net model [19] trained on Kubric [30] and transferred zero-shot.”
Scene Representation Transformer: Geometry-Free Novel View Synthesis Through Set-Latent Scene Representations (2021, background): “We therefore propose this significantly more challenging dataset rendered using photorealistic ray tracing [11].”