# O2O-Afford: Annotation-Free Large-Scale Object-Object Affordance Learning

Source: [O2O-Afford: Annotation-Free Large-Scale Object-Object Affordance Learning](<https://www.sota2.com/research/paper/68fa6b1c1e209131cea2c475>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2025-10-25T23:40:16.064Z

Authors: Kaichun Mo, Yuzhe Qin, Fanbo Xiang, Hao Su, Leonidas Guibas

Published: 2021-06-29T04:38:12.000Z

## Abstract

Contrary to the vast literature in modeling, perceiving, and understanding agent-object (e.g., human-object, hand-object, robot-object) interaction in computer vision and robotics, very few past works have studied the task of object-object interaction, which also plays an important role in robotic manipulation and planning tasks. There is a rich space of object-object interaction scenarios in our daily life, such as placing an object on a messy tabletop, fitting an object inside a drawer, pushing an object using a tool, etc. In this paper, we propose a unified affordance learning framework to learn object-object interaction for various tasks. By constructing four object-object interaction task environments using physical simulation (SAPIEN) and thousands of ShapeNet models with rich geometric diversity, we are able to conduct large-scale object-object affordance learning without the need for human annotations or demonstrations. At the core of technical contribution, we propose an object-kernel point convolution network to reason about detailed interaction between two objects. Experiments on large-scale synthetic data and real-world data prove the effectiveness of the proposed approach. Please refer to the project webpage for code, data, video, and more materials: https://cs.stanford.edu/~kaichun/o2oafford

## Recorded benchmark results

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

- [Pulling on SAPIEN (test)](<https://www.sota2.com/research/sota/pulling-on-sapien-test.md>): pulling; SAPIEN (test); Sample Manipulation Accuracy: 29.93.

- [Pulling Affordance Prediction on SAPIEN PartNet-Mobility & ShapeNet (test)](<https://www.sota2.com/research/sota/pulling-affordance-prediction-on-sapien-partnet-mobility-and-shapenet-test.md>): Pulling Affordance Prediction; SAPIEN PartNet-Mobility & ShapeNet (test); F-Score: 46.03.

- [Pulling Affordance Prediction on SAPIEN (PartNet-Mobility & ShapeNet) Novel (test)](<https://www.sota2.com/research/sota/pulling-affordance-prediction-on-sapien-partnet-mobility-and-shapenet-novel-test.md>): Pulling Affordance Prediction; SAPIEN (PartNet-Mobility & ShapeNet) Novel (test); F-Score: 43.42.

- [Pushing Affordance Prediction on SAPIEN PartNet-Mobility & ShapeNet (test)](<https://www.sota2.com/research/sota/pushing-affordance-prediction-on-sapien-partnet-mobility-and-shapenet-test.md>): Pushing Affordance Prediction; SAPIEN PartNet-Mobility & ShapeNet (test); F-Score: 65.18.

- [Object Affordance on Ego-Exo4D & GIMO (test)](<https://www.sota2.com/research/sota/object-affordance-on-ego-exo4d-and-gimo-test.md>): Object Affordance; Ego-Exo4D & GIMO (test); AUC: 71.52.

- [pushing on SAPIEN (test)](<https://www.sota2.com/research/sota/pushing-on-sapien-test.md>): pushing; SAPIEN (test); Sample Manipulation Accuracy: 34.6.

- [Pushing Affordance Prediction on SAPIEN (PartNet-Mobility & ShapeNet) Novel (test)](<https://www.sota2.com/research/sota/pushing-affordance-prediction-on-sapien-partnet-mobility-and-shapenet-novel-test.md>): Pushing Affordance Prediction; SAPIEN (PartNet-Mobility & ShapeNet) Novel (test); F-Score: 59.04.

## Original sources and implementation

[Original paper](<https://arxiv.org/abs/2106.15087>)

[PDF](<https://arxiv.org/pdf/2106.15087.pdf>)