no code implementations • 2 Feb 2025 • Ye Mao, Weixun Luo, Junpeng Jing, Anlan Qiu, Krystian Mikolajczyk
The rise of vision-language foundation models marks an advancement in bridging the gap between human and machine capabilities in 3D scene reasoning.
no code implementations • 30 Sep 2024 • Junpeng Jing, Ye Mao, Anlan Qiu, Krystian Mikolajczyk
Regarding datasets, current synthetic object-based and indoor datasets are commonly used for training and benchmarking, with a lack of outdoor nature scenarios.
1 code implementation • 25 Apr 2024 • Ye Mao, Junpeng Jing, Krystian Mikolajczyk
In this paper, we present OpenDlign, a novel open-world 3D model using depth-aligned images generated from a diffusion model for robust multimodal alignment.
Ranked #1 on
Zero-shot 3D classification
on OmniObject3D (Pretrained on ShapeNet)
(using extra training data)
no code implementations • 16 Mar 2024 • Junpeng Jing, Ye Mao, Krystian Mikolajczyk
Towards this challenge, we develop a bidirectional alignment mechanism for adjacent frames as a fundamental operation.
no code implementations • ICCV 2023 • Junpeng Jing, Jiankun Li, Pengfei Xiong, Jiangyu Liu, Shuaicheng Liu, Yichen Guo, Xin Deng, Mai Xu, Lai Jiang, Leonid Sigal
A novel Uncertainty Guided Adaptive Correlation (UGAC) module is introduced to robustly adapt the same model for different scenarios.
1 code implementation • ICCV 2021 • Junpeng Jing, Xin Deng, Mai Xu, Jianyi Wang, Zhenyu Guan
Capacity, invisibility and security are three primary challenges in image hiding task.