9 papers
MoECodec: Image Compression for joint human and machine perception via Mixture-of-Experts
Jiancheng Zhao, Xiang Ji, Yifan Zhan +2
Image compression for machines calls for a unified codec that serves multiple downstream vision tasks. Existing approaches either adopt task-specific end-to-end designs, raising pa…
Moment-Reenacting: Inverse Motion Degradation with Cross-shutter Guidance
Xiang Ji, Guixu Lin, Zhengwei Yin +2
Motion degradation, manifested as blur in global shutter (GS) images or rolling shutter (RS) distortion in RS counterparts, remains a fundamental challenge in computational imaging…
AniCrafter: Customizing Realistic Human-Centric Animation via Avatar-Background Conditioning in Video Diffusion Models
Muyao Niu, Mingdeng Cao, Yifan Zhan +6
Recent advances in video diffusion models have substantially enhanced character animation techniques. However, existing methods primarily depend on structural conditions, such as D…
Tree-NeRV: A Tree-Structured Neural Representation for Efficient Non-Uniform Video Encoding
Jiancheng Zhao, Yifan Zhan, Qingtian Zhu +5
Implicit Neural Representations for Videos (NeRV) have emerged as a powerful paradigm for video representation, enabling direct mappings from frame indices to video frames. However…
All-in-One Transferring Image Compression from Human Perception to Multi-Machine Perception
Jiancheng Zhao, Xiang Ji, Yinqiang Zheng
Efficiently transferring Learned Image Compression (LIC) model from human perception to machine perception is an emerging challenge in vision-centric representation learning. Exist…
MSPLoRA: A Multi-Scale Pyramid Low-Rank Adaptation for Efficient Model Fine-Tuning
Jiancheng Zhao, Xingda Yu, Zhen Yang
Parameter-Efficient Fine-Tuning (PEFT) has become an essential approach for adapting large-scale pre-trained models while reducing computational costs. Among PEFT methods, LoRA sig…
LoR2C : Low-Rank Residual Connection Adaptation for Parameter-Efficient Fine-Tuning
Jiancheng Zhao, Xingda Yu, Yuxiang Zhang +1
In recent years, pretrained large language models have demonstrated outstanding performance across various natural language processing tasks. However, full-parameter fine-tuning me…
ToMiE: Towards Explicit Exoskeleton for the Reconstruction of Complicated 3D Human Avatars
Yifan Zhan, Qingtian Zhu, Muyao Niu +6
In this paper, we highlight a critical yet often overlooked factor in most 3D human tasks, namely modeling complicated 3D human with with hand-held objects or loose-fitting clothin…
MoEC: Mixture of Experts Implicit Neural Compression
Jianchen Zhao, Cheng-Ching Tseng, Ming Lu +4
Emerging Implicit Neural Representation (INR) is a promising data compression technique, which represents the data using the parameters of a Deep Neural Network (DNN). Existing met…