Contact Sales
EgoChores Dataset

A robot-grade egocentric dataset for household manipulation.

First-person video of real household chores, capturing the dexterous hand movements robots need to learn manipulation. License it off the shelf as raw video, annotated with our standard schema, or with fully custom annotation.

Request the free sample

EgoChores Dataset Consultation

Four first-person frames of household chores from the EgoChores dataset
The focus

EgoChores is a targeted, high-density egocentric video dataset of everyday household cleaning and tidying, captured from a head-mounted first-person perspective with a focus on dexterous hand-object manipulation. Our goal was to address something we found lacking in currently available datasets: a single, high-demand task family, available off the shelf, with production-grade annotations optimized specifically for pre-training embodied AI.

What's included

Dataset specifications

Covered activities

  • Vacuuming & mopping
  • Sweeping
  • Wiping countertops & flat surfaces
  • Scrubbing stovetop
  • Cleaning sink
  • Wiping shower/tub
Plus 21 more Show less
  • Cleaning mirrors/glass
  • Dusting shelves & furniture
  • Dusting ceiling fans
  • Cleaning microwave interior
  • Cleaning oven interior
  • Cleaning refrigerator shelves
  • Loading dishwasher
  • Unloading dishwasher
  • Hand washing dishes
  • Wiping appliance exteriors & handles
  • Trash — removal & bag replacement
  • Scrubbing toilet bowl
  • Scrubbing bathroom tiles
  • Cleaning drain/trap
  • Folding laundry
  • Organizing cabinets
  • Cleaning windows (interior)
  • Spot cleaning carpet
  • Making bed / changing linens
  • Wiping baseboards & trim
  • Picking up pet waste (outdoors)

Capture protocol

  • Head-mounted POV (1080p, 60fps)
  • Natural, un-staged home environments
  • Geographically diverse households
  • Diverse participant demographics
  • Continuous, unedited task recordings

Annotation schema

  • Per-frame action segmentation
  • Dexterous hand-object interaction tracking
  • Strict temporal task boundaries
  • Hierarchical object category labels
  • Standardized, custom taxonomy
Architectural deep dive

A published, machine-readable annotation schema

Precisely annotated in Label Studio, every clip is segmented into timed actions and resolved against a hierarchical object taxonomy created under the supervision of robotics researchers. Commission custom collection runs in the same schema to extend and adapt the dataset.

Timeline & taxonomy
[0:00 – 0:02] Action: Open Appliance Hand: Right Object: Dishwasher Door
[0:03 – 0:05] Action: Grasp Object Hand: Left Object: Ceramic Plate
[0:06 – 0:09] Action: Insert Object Hand: Left Object: Dishwasher Lower Rack
The object hierarchy
  • Environment: Kitchen
    • Appliance: Dishwasher
      • Component: Lower Rack
        • Prongs
      • Manipuland: Ceramic Plate
        • Rim
Engineering note

EgoChores ships with a fully published, machine-readable annotation schema. Your robotics teams can train on it natively and later use the exact same taxonomy to commission custom collection runs.

Designed for embodied AI

Built for how robot models actually learn

VLA

Vision-Language-Action models

Rich, textual step descriptions aligned precisely to visual frame boundaries.

Policy

Robot policy & imitation learning

Continuous first-person trajectories tracking multi-modal hand states and object interactions.

Planning

Manipulation planning

Clear pre-condition and post-condition states captured before and after every major task boundary.

Fine-tune

Foundation model fine-tuning

High-density, domain-specific data to bridge the simulation-to-reality (Sim2Real) gap.

Access & engagement

Three ways to license EgoChores off the shelf

License the dataset at the level of annotation your team needs — or request a free sample to evaluate it first.

Raw Video Data

What it is

The full EgoChores video corpus with basic clip-level tags and metadata.

Best for

Teams running their own annotation pipeline, or pre-training directly on raw egocentric video.

EgoChores — Raw Video Data

Standard Annotation

What it is

EgoChores video with the HumanSignal standard label schema applied — action segmentation, dexterous hand-object interaction, and object taxonomy.

Best for

Immediate integration into VLA training loops and manipulation benchmarking.

EgoChores — Standard Annotation

Custom Annotation

What it is

EgoChores video with heavier, bespoke annotation — custom object classes, edge cases, and proprietary schema configurations.

Best for

Production-grade robotics labs scaling toward deployment.

EgoChores — Custom Annotation

Documentation & resources

Everything your team needs to evaluate and integrate

FAQ

The technical questions, answered

How does EgoChores differ from massive open-world datasets like Ego4D?

While Ego4D provides unparalleled unconstrained scale, it lacks the dense task-family specificity needed for highly repeatable robotic policy learning. EgoChores provides tight, standardized taxonomies across identical task categories, minimizing background noise and maximizing target interaction density.

Can HumanSignal reproduce this exact capture protocol in other regions or countries?

Yes. Our human data collection networks span multiple global regions, allowing us to capture diverse architectural styles, appliance form factors, and cultural variations in chore performance.

What annotation formats do you support out of the box?

EgoChores ships natively in standard JSON / COCO-style formats for vision models. Our internal annotation tooling also supports custom schema compilation to format outputs directly for your team's proprietary data loaders.

Ready to build data pipelines your competitors can't replicate?

Let's design your proprietary dataset program today.

EgoChores Dataset — Data Services