WayPOP: A Panoramic Open-Set Panoptic Tracking Benchmark
WayPOP: A Panoramic Open-Set Panoptic Tracking Benchmark
This research introduces WayPOP, a new benchmark for panoramic open-set panoptic tracking. Built on real Waymo PVPS sequences augmented with diverse out-of-taxonomy 3D assets from Objaverse, the dataset provides temporally and cross-view consistent annotations for unknown objects. It contains approximately 670 curated assets instantiated as 980 unique unknown-object tracks, yielding 53,452 mask annotations across 21,300 camera images. The authors evaluate representative open-set and open-vocabulary segmentation methods, revealing substantial limitations in current approaches for robust unknown-object localization and identity association across time and views.
The paper introduces WayPOP, a benchmark dataset comprising approximately 670 curated assets and 53,452 mask annotations across 21,300 images to evaluate open-set panoptic tracking in panoramic views.
Source: arXiv Robotics — research abstracts · Read original article ↗
Loading article text…
What the source reports
Publisher-reported claims, with original evidence. These results have not been independently verified by RoboSignal.
Reported numbers
camera images
21,300
View original evidence
nd surrounding traffic move, providing temporally and cross-view consistent panoptic annotations and auxiliary LiDAR range observations. WayPOP contains approximately 670 curated assets instantiated as 980 unique unknown-object tracks, yielding 53,452 unknown-object mask annotations across 21,300 camera images. Our evaluation protocol defines a hierarchy of tasks from view-independent open-set panoptic segmentation t
Open source S4unknown-object tracks
980
View original evidence
nd surrounding traffic move, providing temporally and cross-view consistent panoptic annotations and auxiliary LiDAR range observations. WayPOP contains approximately 670 curated assets instantiated as 980 unique unknown-object tracks, yielding 53,452 unknown-object mask annotations across 21,300 camera images. Our evaluation protocol defines a hierarchy of tasks from view-independent open-set panoptic segmentation t
Open source S4unknown-object mask annotations
53,452
View original evidence
nd surrounding traffic move, providing temporally and cross-view consistent panoptic annotations and auxiliary LiDAR range observations. WayPOP contains approximately 670 curated assets instantiated as 980 unique unknown-object tracks, yielding 53,452 unknown-object mask annotations across 21,300 camera images. Our evaluation protocol defines a hierarchy of tasks from view-independent open-set panoptic segmentation t
Open source S4curated assets
≈670
View original evidence
nd surrounding traffic move, providing temporally and cross-view consistent panoptic annotations and auxiliary LiDAR range observations. WayPOP contains approximately 670 curated assets instantiated as 980 unique unknown-object tracks, yielding 53,452 unknown-object mask annotations across 21,300 camera images. Our evaluation protocol defines a hierarchy of tasks from view-independent open-set panoptic segmentation t
Open source S4
- paper: not_reported
What remains unknown
Dataset licensing and commercial-use rights are not established here.
Source excerpts and review record
Automatically extracted; no manual editorial approval recorded.
nd surrounding traffic move, providing temporally and cross-view consistent panoptic annotations and auxiliary LiDAR range observations. WayPOP contains approximately 670 curated assets instantiated as 980 unique unknown-object tracks, yielding 53,452 unknown-object mask annotations across 21,300 camera images. Our evaluation protocol defines a hierarchy of tasks from view-independent open-set panoptic segmentation t
Open source S4
Implications for data suppliers
RoboSignal interpretation and collection questions, not statements of buyer demand.
- Confirm the required data type and collection setting with the buyer; this source does not establish a complete collection specification.
- Compare the reported units and scope before using these quantities in a budget. Recording hours, sensor-hours and trajectories are different measures.
- Validate demand and acceptance criteria with a buyer before scaling. Publication, popularity and a research result do not establish a purchase commitment.
Source：arXiv Robotics — research abstracts · arxiv.org