TartanDrive is selected as an outstanding learning paper finalist in ICRA2022

TartanDrive is selected as an outstanding learning paper finalist in ICRA2022

Published: by

TartanDrive is selected as an outstanding learning paper finalist in ICRA2022

We took an all-terrain vehicle on wild rides through tall grass, loose gravel and mud to gather data about how the ATV interacted with a challenging, off-road environment.

The dataset and the paper are openly available. Read more about it in our linked article below:

Latest Research

Co-Me: Confidence-Guided Token Merging for Visual Geometry Transformer
Co-Me: Confidence-Guided Token Merging for Visual Geometry Transformer

Co-Me is an acceleration mechanism that uses a lightweight confidence predictor to selectively me...

Published: May 18, 2026

Unified Spherical Frontend: Learning Rotation-Equivariant Representations of Spherical Images from Any Camera
Unified Spherical Frontend: Learning Rotation-Equivariant Representations of Spherical Images from Any Camera

A distortion-free lens-agnostic rotation-equivariant vision framework for modern perception

Published: Apr 3, 2026

RADSeg: Unleashing Parameter and Compute Efficient Zero-Shot Open-Vocabulary Segmentation Using Agglomerative Models
RADSeg: Unleashing Parameter and Compute Efficient Zero-Shot Open-Vocabulary Segmentation Using Agglomerative Models

RADSeg is a dense, language-aligned feature encoder that enables low-parameter, low-latency open-...