no code implementations • 25 Apr 2024 • Chenxi Liu, Gan Sun, Wenqi Liang, Jiahua Dong, Can Qin, Yang Cong
To deal with catastrophic forgetting amongst past learned styles, we devise a dual regularization for shared-LoRA module to optimize the direction of model update, which could regularize the diffusion model from both weight and feature aspects, respectively.
1 code implementation • 17 Mar 2024 • Guohao Sun, Can Qin, Jiamian Wang, Zeyuan Chen, ran Xu, Zhiqiang Tao
Recent advancements in the vision-language model have shown notable generalization in vision-language tasks after visual instruction tuning.
Ranked #50 on Visual Question Answering on MM-Vet
2 code implementations • 11 Dec 2023 • Jiaming Liu, Yue Wu, Maoguo Gong, Qiguang Miao, Wenping Ma, Can Qin
3D Single Object Tracking (SOT) stands a forefront task of computer vision, proving essential for applications like autonomous driving.
no code implementations • 13 Aug 2023 • Haichao Zhang, Can Qin, Yu Yin, Yun Fu
This approach can serve as a plug-and-play data generation and augmentation module for existing camouflaged object detection tasks and provides a novel way to introduce more diversity and distributions into current camouflage datasets.
1 code implementation • NeurIPS 2023 • Can Qin, Shu Zhang, Ning Yu, Yihao Feng, Xinyi Yang, Yingbo Zhou, Huan Wang, Juan Carlos Niebles, Caiming Xiong, Silvio Savarese, Stefano Ermon, Yun Fu, ran Xu
Visual generative foundation models such as Stable Diffusion show promise in navigating these goals, especially when prompted with arbitrary languages.
no code implementations • CVPR 2023 • Vibashan VS, Ning Yu, Chen Xing, Can Qin, Mingfei Gao, Juan Carlos Niebles, Vishal M. Patel, ran Xu
In summary, an OV method learns task-specific information using strong supervision from base annotations and novel category information using weak supervision from image-captions pairs.
1 code implementation • ICCV 2023 • Can Qin, Ning Yu, Chen Xing, Shu Zhang, Zeyuan Chen, Stefano Ermon, Yun Fu, Caiming Xiong, ran Xu
Empirical results show that GlueNet can be trained efficiently and enables various capabilities beyond previous state-of-the-art models: 1) multilingual language models such as XLM-Roberta can be aligned with existing T2I models, allowing for the generation of high-quality images from captions beyond English; 2) GlueNet can align multi-modal encoders such as AudioCLIP with the Stable Diffusion model, enabling sound-to-image generation; 3) it can also upgrade the current text encoder of the latent diffusion model for challenging case generation.
1 code implementation • 16 Mar 2023 • Shu Zhang, Xinyi Yang, Yihao Feng, Can Qin, Chia-Chih Chen, Ning Yu, Zeyuan Chen, Huan Wang, Silvio Savarese, Stefano Ermon, Caiming Xiong, ran Xu
Incorporating human feedback has been shown to be crucial to align text generated by large language models to human preferences.
2 code implementations • 2 Mar 2023 • Xu Ma, Yuqian Zhou, Huan Wang, Can Qin, Bin Sun, Chang Liu, Yun Fu
Context clusters (CoCs) view an image as a set of unorganized points and extract features via simplified clustering algorithm.
1 code implementation • 28 Jan 2023 • Yizhou Wang, Can Qin, Yue Bai, Yi Xu, Xu Ma, Yun Fu
With the same perturbation magnitude, the testing reconstruction error of the normal frames lowers more than that of the abnormal frames, which contributes to mitigating the overfitting problem of reconstruction.
2 code implementations • 12 Jan 2023 • Huan Wang, Can Qin, Yue Bai, Yun Fu
The state of neural network pruning has been noticed to be unclear and even confusing for a while, largely due to "a lack of standardized benchmarks and metrics" [3].
1 code implementation • 23 Dec 2022 • Xu Ma, Huan Wang, Can Qin, Kunpeng Li, Xingchen Zhao, Jie Fu, Yun Fu
Vision Transformers have shown great promise recently for many vision tasks due to the insightful architecture design and attention mechanism.
1 code implementation • ICLR 2022 • Xu Ma, Can Qin, Haoxuan You, Haoxi Ran, Yun Fu
We notice that detailed local geometrical information probably is not the key to point cloud analysis -- we introduce a pure residual MLP network, called PointMLP, which integrates no sophisticated local geometrical extractors but still performs very competitively.
Ranked #4 on Point Cloud Segmentation on PointCloud-C
Point Cloud Segmentation Supervised Only 3D Point Cloud Classification
1 code implementation • 12 Dec 2021 • Can Qin, Lichen Wang, Qianqian Ma, Yu Yin, Huan Wang, Yun Fu
Semi-supervised domain adaptation (SSDA) is quite a challenging problem requiring methods to overcome both 1) overfitting towards poorly annotated data and 2) distribution shift across domains.
1 code implementation • NeurIPS 2021 • Can Qin, Handong Zhao, Lichen Wang, Huan Wang, Yulun Zhang, Yun Fu
For slow learning of graph similarity, this paper proposes a novel early-fusion approach by designing a co-attention-based feature fusion network on multilevel GNN features.
1 code implementation • NeurIPS 2021 • Yulun Zhang, Huan Wang, Can Qin, Yun Fu
To address the above issues, we propose aligned structured sparsity learning (ASSL), which introduces a weight normalization layer and applies $L_2$ regularization to the scale parameters for sparsity.
1 code implementation • 25 Nov 2021 • Yizhou Wang, Can Qin, Rongzhe Wei, Yi Xu, Yue Bai, Yun Fu
Next we add adversarial perturbation to the transformed features to decrease their softmax scores of the predicted labels and design anomaly scores based on the predictive uncertainties of the classifier on these perturbed features.
1 code implementation • 31 Oct 2021 • Joseph P. Robinson, Can Qin, Ming Shao, Matthew A. Turk, Rama Chellappa, Yun Fu
Recognizing Families In the Wild (RFIW), held as a data challenge in conjunction with the 16th IEEE International Conference on Automatic Face and Gesture Recognition (FG), is a large-scale, multi-track visual kinship recognition evaluation.
no code implementations • 29 Sep 2021 • Yi Xu, Lichen Wang, Yizhou Wang, Can Qin, Yulun Zhang, Yun Fu
In this paper, we propose a novel framework, MemREIN, which considers Memorized, Restitution, and Instance Normalization for cross-domain few-shot learning.
no code implementations • ICLR 2022 • Yulun Zhang, Huan Wang, Can Qin, Yun Fu
Specifically, for the layers connected by the same residual, we select the filters of the same indices as unimportant filters.
no code implementations • 29 Sep 2021 • Huan Wang, Can Qin, Yue Bai, Yun Fu
Several recent works questioned the value of inheriting weight in structured neural network pruning because they empirically found training from scratch can match or even outperform finetuning a pruned model.
1 code implementation • 22 Jun 2021 • Yizhou Wang, Yue Kang, Can Qin, Huan Wang, Yi Xu, Yulun Zhang, Yun Fu
The intuition is that gradient with momentum contains more accurate directional information and therefore its second moment estimation is a more favorable option for learning rate scaling than that of the raw gradient.
no code implementations • 12 May 2021 • Huan Wang, Can Qin, Yue Bai, Yun Fu
This paper is meant to explain it through the lens of dynamical isometry [42].
1 code implementation • 16 Mar 2021 • Joseph P Robinson, Can Qin, Yann Henon, Samson Timoner, Yun Fu
This scheme boosts the average performance and preserves identity information while removing demographic knowledge.
2 code implementations • 11 Mar 2021 • Huan Wang, Can Qin, Yue Bai, Yulun Zhang, Yun Fu
Neural network pruning typically removes connections or neurons from a pretrained converged model; while a new pruning paradigm, pruning at initialization (PaI), attempts to prune a randomly initialized network.
no code implementations • ICCV 2021 • Yulun Zhang, Donglai Wei, Can Qin, Huan Wang, Hanspeter Pfister, Yun Fu
However, the basic convolutional layer in CNNs is designed to extract local patterns, lacking the ability to model global context.
no code implementations • 1 Jan 2021 • Lichen Wang, Bo Zong, Yunyu Liu, Can Qin, Wei Cheng, Wenchao Yu, Xuchao Zhang, Haifeng Chen, Yun Fu
As texts always contain a large proportion of task-irrelevant words, accurate alignment between aspects and their sentimental descriptions is the most crucial and challenging step.
1 code implementation • ICLR 2021 • Huan Wang, Can Qin, Yulun Zhang, Yun Fu
Regularization has long been utilized to learn sparsity in deep neural network pruning.
no code implementations • 7 Dec 2020 • Yu Yin, Joseph P. Robinson, Songyao Jiang, Yue Bai, Can Qin, Yun Fu
Even as impressive milestones are achieved in synthesizing faces, the importance of preserving identity is needed in practice and should not be overlooked.
1 code implementation • 16 Feb 2020 • Joseph P. Robinson, Gennady Livitz, Yann Henon, Can Qin, Yun Fu, Samson Timoner
Thus, the conventional approach of learning a global threshold for all pairs resulting in performance gaps among subgroups.
1 code implementation • 6 Feb 2020 • Can Qin, Lichen Wang, Qianqian Ma, Yu Yin, Huan Wang, Yun Fu
Current adversarial adaptation methods attempt to align the cross-domain features, whereas two challenges remain unsolved: 1) the conditional distribution mismatch and 2) the bias of the decision boundary towards the source domain.
1 code implementation • 4 Feb 2020 • Changyu Deng, Yizhou Wang, Can Qin, Yun Fu, Wei Lu
A small number of training data is generated dynamically based on the DNN's prediction of the optimum.
2 code implementations • NeurIPS 2019 • Can Qin, Haoxuan You, Lichen Wang, C. -C. Jay Kuo, Yun Fu
Specifically, most general-purpose DA methods that struggle for global feature alignment and ignore local geometric information are not suitable for 3D domain alignment.
Ranked #1 on Unsupervised Domain Adaptation on PreSIL to KITTI