Current visual foundation models are trained purely on unstructured 2D data, limiting their
understanding of 3D structure of objects and scenes. In this work, we show that fine-tuning …

This work addresses the challenge of high-quality surface normal estimation from monocular
colored inputs (ie, images and videos), a field which has recently been revolutionized by …

This paper introduces a novel benchmark for efficient image upscaling as part of the AIS
2024 Real-Time Image Super-Resolution (RTSR) Challenge which aims to upscale …

We present a simple self-supervised method to enhance the performance of ViT features for
dense downstream tasks. Our Lightweight Feature Transform (LiFT) is a straightforward and …

Remote sensing image plays an irreplaceable role in fields such as agriculture, water
resources, military, and disaster relief. Pixel-level interpretation is a critical aspect of remote …

In multimodal large language models (MLLMs), vision transformers (ViTs) are widely
employed for visual encoding. However, their performance in solving universal MLLM tasks …

While recent foundational video generators produce visually rich output, they still struggle
with appearance drift, where objects gradually degrade or change inconsistently across …

Generalization to novel object configurations and instances across diverse tasks and
environments is a critical challenge in robotics. Keypoint-based representations have been …

3D part segmentation is a crucial and challenging task in 3D perception, playing a vital role
in applications such as robotics, 3D generation, and 3D editing. Recent methods harness …

Feature upsampling is a fundamental and indispensable ingredient of almost all current
network structures for image segmentation tasks. Recently, a popular similarity-based …