pov
Datasets
All datasets matching “pov”tracker-pov
Eidon Tracker POV
1,274 hours of egocentric video paired with 7-point IMU arm tracking, recorded during ordinary household work.
Contributors wore a head-mounted camera and a seven-sensor IMU harness while doing real chores in their own homes: laundry, cleaning, dishes, cooking. Each recording pairs first-person video with 24 Hz orientation data for both hands, both forearms, both upper arms, and the chest.
This is a complete, final release. Eidon AI (Solidic Labs Inc) has wound… See the full description on the dataset page: https://huggingface.co/datasets/eidon-ai/tracker-pov.tracker-pov-imu
Eidon Tracker POV: IMU
24 Hz orientation and motion data from a seven-point IMU harness, paired with the egocentric
video in eidon-ai/tracker-pov.
One row per (recording, timestamp, body slot), roughly 780 million rows. Join to the video
metadata on recording_id.
This repo holds the sensor data only. There is no video here.
The release sits in three places:
Contents
Size
tracker-pov
the 13,451 MP4s and metadata.parquet
9.05 TB
this repo… See the full description on the dataset page: https://huggingface.co/datasets/eidon-ai/tracker-pov-imu.PovertyMap
PovertyMap-Wilds: Poverty mapping across different countries
Description
This is a processed version of LandSat 5/7/8 satellite imagery originally from Google Earth Engine under the names LANDSAT/LC08/C01/T1_SR,LANDSAT/LE07/C01/T1_SR,LANDSAT/LT05/C01/T1_SR,
nighttime light imagery from the DMSP and VIIRS satellites (Google Earth Engine names NOAA/DMSP-OLS/CALIBRATED_LIGHTS_V4 and NOAA/VIIRS/DNB/MONTHLY_V1/VCMSLCFG)
and processed DHS survey metadata obtained from… See the full description on the dataset page: https://huggingface.co/datasets/1aurent/PovertyMap.pov-data
POV Data
Public de-identified first-person recordings, captured on Project Aria glasses
and covering 2026-04-15 through 2026-08-18.
Recordings live at aria/YYYY-MM-DD/*.tar — uncompressed, one tar per recording.
What's in a recording
<recording>/
├── picture/masked/ RGB frames, 2880×2880, ~1 fps, faces and
│ identifying screen content already masked
├── picture/ocr_text/ text recognised in each frame
├── eye_tracking/
│ ├──… See the full description on the dataset page: https://huggingface.co/datasets/mmm8383/pov-data.POVBench
POVBench
Contextual Observer Grounding: Evaluating Situated Spatial Reasoning in Vision-Language ModelsEMNLP 2026 Findings
Project page · Code
Given a sentence in which an observer says where they last saw an object — from
their own point of view — a model must recover that perspective and localize the
target in image space.
Crucially, the observer's right is not necessarily aligned with the camera's right.
Three conditions progressively reduce the amount of reasoning required:… See the full description on the dataset page: https://huggingface.co/datasets/owl-owl/POVBench.povqa-preprocessed
POVQA Preprocessed Frames and Subtitle-Frame Alignments
Accepted to the MAR Workshop at CVPR 2026.
Paper |
Preprocessed Dataset |
Project Website
This dataset release contains the **preprocessed artifacts** used by POVQA. It is
intended to support research reproducibility without redistributing raw source
videos or raw subtitle files.
What is included
Preprocessed frame bundles for 12 movies and 4 pooling variants
KEY_FRAMES bundles where available… See the full description on the dataset page: https://huggingface.co/datasets/ashimdahal/povqa-preprocessed.