Edge-Assisted Lightweight Region-of-Interest Extraction and Transmission for Vehicle Perception

Kavli Affiliate: Cheng Peng | First 5 Authors: Yan Cheng, Peng Yang, Ning Zhang, Jiawei Hou, | Summary: To enhance on-road environmental perception for autonomous driving, accurate and real-time analytics on high-resolution video frames generated from on-board cameras be-comes crucial. In this paper, we design a lightweight object location method based on class activation mapping […]


Continue.. Edge-Assisted Lightweight Region-of-Interest Extraction and Transmission for Vehicle Perception

Non-equilibrium spin polarisation via interfacial exchange-field spin filtering

Kavli Affiliate: Zheng Zhu | First 5 Authors: Prasanta Muduli, NaĆ«mi Leo, Mingran Xu, Zheng Zhu, Jorge Puebla | Summary: A key phenomenon that enables nanoscale spintronic devices is the efficient inter-conversion between spin and charge degrees of freedom. Here, we experimentally demonstrate a pathway to generate current-induced spin polarization at the interface between an […]


Continue.. Non-equilibrium spin polarisation via interfacial exchange-field spin filtering

Unified Single-Stage Transformer Network for Efficient RGB-T Tracking

Kavli Affiliate: Zheng Zhu | First 5 Authors: Jianqiang Xia, DianXi Shi, Ke Song, Linna Song, XiaoLei Wang | Summary: Most existing RGB-T tracking networks extract modality features in a separate manner, which lacks interaction and mutual guidance between modalities. This limits the network’s ability to adapt to the diverse dual-modality appearances of targets and […]


Continue.. Unified Single-Stage Transformer Network for Efficient RGB-T Tracking

Project Aria: A New Tool for Egocentric Multi-Modal AI Research

Kavli Affiliate: Cheng Peng | First 5 Authors: Jakob Engel, Kiran Somasundaram, Michael Goesele, Albert Sun, Alexander Gamino | Summary: Egocentric, multi-modal data as available on future augmented reality (AR) devices provides unique challenges and opportunities for machine perception. These future devices will need to be all-day wearable in a socially acceptable form-factor to support […]


Continue.. Project Aria: A New Tool for Egocentric Multi-Modal AI Research

LISTER: Neighbor Decoding for Length-Insensitive Scene Text Recognition

Kavli Affiliate: Cheng Peng | First 5 Authors: Changxu Cheng, Peng Wang, Cheng Da, Qi Zheng, Cong Yao | Summary: The diversity in length constitutes a significant characteristic of text. Due to the long-tail distribution of text lengths, most existing methods for scene text recognition (STR) only work well on short or seen-length text, lacking […]


Continue.. LISTER: Neighbor Decoding for Length-Insensitive Scene Text Recognition

GRIP: Generating Interaction Poses Using Latent Consistency and Spatial Cues

Kavli Affiliate: Yi Zhou | First 5 Authors: Omid Taheri, Yi Zhou, Dimitrios Tzionas, Yang Zhou, Duygu Ceylan | Summary: Hands are dexterous and highly versatile manipulators that are central to how humans interact with objects and their environment. Consequently, modeling realistic hand-object interactions, including the subtle motion of individual fingers, is critical for applications […]


Continue.. GRIP: Generating Interaction Poses Using Latent Consistency and Spatial Cues

COCA: Classifier-Oriented Calibration for Source-Free Universal Domain Adaptation via Textual Prototype

Kavli Affiliate: Yi Zhou | First 5 Authors: Xinghong Liu, Yi Zhou, Tao Zhou, Chun-Mei Feng, Ling Shao | Summary: Universal Domain Adaptation (UniDA) aims to distinguish common and private classes between the source and target domains where domain shift exists. Recently, due to more stringent data restrictions, researchers have introduced Source-Free UniDA (SF-UniDA) in […]


Continue.. COCA: Classifier-Oriented Calibration for Source-Free Universal Domain Adaptation via Textual Prototype

COCA: Classifier-Oriented Calibration via Textual Prototype for Source-Free Universal Domain Adaptation

Kavli Affiliate: Yi Zhou | First 5 Authors: Xinghong Liu, Yi Zhou, Tao Zhou, Chun-Mei Feng, Ling Shao | Summary: Universal domain adaptation (UniDA) aims to address domain and category shifts across data sources. Recently, due to more stringent data restrictions, researchers have introduced source-free UniDA (SF-UniDA). SF-UniDA methods eliminate the need for direct access […]


Continue.. COCA: Classifier-Oriented Calibration via Textual Prototype for Source-Free Universal Domain Adaptation

COCA: Classifier-Oriented Calibration via Textual Prototype for Source-Free Universal Domain Adaptation

Kavli Affiliate: Yi Zhou | First 5 Authors: Xinghong Liu, Yi Zhou, Tao Zhou, Chun-Mei Feng, Ling Shao | Summary: Universal domain adaptation (UniDA) aims to address domain and category shifts across data sources. Recently, due to more stringent data restrictions, researchers have introduced source-free UniDA (SF-UniDA). SF-UniDA methods eliminate the need for direct access […]


Continue.. COCA: Classifier-Oriented Calibration via Textual Prototype for Source-Free Universal Domain Adaptation

The effect of the adiabatic assumption on asteroseismic scaling relations for luminous red giants

Kavli Affiliate: Lars Bildsten | First 5 Authors: Joel C. Zinn, Marc H. Pinsonneault, Lars Bildsten, Dennis Stello, | Summary: Although stellar radii from asteroseismic scaling relations agree at the percent level with independent estimates for main sequence and most first-ascent red giant branch stars, the scaling relations over-predict radii at the tens of percent […]


Continue.. The effect of the adiabatic assumption on asteroseismic scaling relations for luminous red giants