CVPR 2023 Open Access Repository

The 6th Efficient Deep Learning for Computer Vision

ETAD: Training Action Detection End to End on a Laptop: Shuming Liu,

Mengmeng Xu,

Chen Zhao,

Xu Zhao,

Bernard Ghanem; [pdf] [supp] [arXiv]
[bibtex]
@InProceedings{Liu_2023_CVPR, author = {Liu, Shuming and Xu, Mengmeng and Zhao, Chen and Zhao, Xu and Ghanem, Bernard}, title = {ETAD: Training Action Detection End to End on a Laptop}, booktitle = {Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) Workshops}, month = {June}, year = {2023}, pages = {4525-4534} }
STAR: Sparse Thresholded Activation Under Partial-Regularization for Activation Sparsity Exploration: Zeqi Zhu,

Arash Pourtaherian,

Luc Waeijen,

Egor Bondarev,

Orlando Moreira; [pdf] [supp]
[bibtex]
@InProceedings{Zhu_2023_CVPR, author = {Zhu, Zeqi and Pourtaherian, Arash and Waeijen, Luc and Bondarev, Egor and Moreira, Orlando}, title = {STAR: Sparse Thresholded Activation Under Partial-Regularization for Activation Sparsity Exploration}, booktitle = {Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) Workshops}, month = {June}, year = {2023}, pages = {4554-4563} }
Data-Free Model Pruning at Initialization via Expanders: James Stewart,

Umberto Michieli,

Mete Ozay; [pdf]
[bibtex]
@InProceedings{Stewart_2023_CVPR, author = {Stewart, James and Michieli, Umberto and Ozay, Mete}, title = {Data-Free Model Pruning at Initialization via Expanders}, booktitle = {Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) Workshops}, month = {June}, year = {2023}, pages = {4519-4524} }
Accelerable Lottery Tickets With the Mixed-Precision Quantization: Zhangheng Li,

Yu Gong,

Zhenyu Zhang,

Xingyun Xue,

Tianlong Chen,

Yi Liang,

Bo Yuan,

Zhangyang Wang; [pdf]
[bibtex]
@InProceedings{Li_2023_CVPR, author = {Li, Zhangheng and Gong, Yu and Zhang, Zhenyu and Xue, Xingyun and Chen, Tianlong and Liang, Yi and Yuan, Bo and Wang, Zhangyang}, title = {Accelerable Lottery Tickets With the Mixed-Precision Quantization}, booktitle = {Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) Workshops}, month = {June}, year = {2023}, pages = {4604-4612} }
BinaryViT: Pushing Binary Vision Transformers Towards Convolutional Models: Phuoc-Hoan Charles Le,

Xinlin Li; [pdf] [supp] [arXiv]
[bibtex]
@InProceedings{Le_2023_CVPR, author = {Le, Phuoc-Hoan Charles and Li, Xinlin}, title = {BinaryViT: Pushing Binary Vision Transformers Towards Convolutional Models}, booktitle = {Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) Workshops}, month = {June}, year = {2023}, pages = {4665-4674} }
Token Merging for Fast Stable Diffusion: Daniel Bolya,

Judy Hoffman; [pdf] [supp] [arXiv]
[bibtex]
@InProceedings{Bolya_2023_CVPR, author = {Bolya, Daniel and Hoffman, Judy}, title = {Token Merging for Fast Stable Diffusion}, booktitle = {Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) Workshops}, month = {June}, year = {2023}, pages = {4599-4603} }
DeCAtt: Efficient Vision Transformers With Decorrelated Attention Heads: Mayukh Bhattacharyya,

Soumitri Chattopadhyay,

Sayan Nag; [pdf]
[bibtex]
@InProceedings{Bhattacharyya_2023_CVPR, author = {Bhattacharyya, Mayukh and Chattopadhyay, Soumitri and Nag, Sayan}, title = {DeCAtt: Efficient Vision Transformers With Decorrelated Attention Heads}, booktitle = {Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) Workshops}, month = {June}, year = {2023}, pages = {4695-4699} }
Speed Is All You Need: On-Device Acceleration of Large Diffusion Models via GPU-Aware Optimizations: Yu-Hui Chen,

Raman Sarokin,

Juhyun Lee,

Jiuqiang Tang,

Chuo-Ling Chang,

Andrei Kulik,

Matthias Grundmann; [pdf] [arXiv]
[bibtex]
@InProceedings{Chen_2023_CVPR, author = {Chen, Yu-Hui and Sarokin, Raman and Lee, Juhyun and Tang, Jiuqiang and Chang, Chuo-Ling and Kulik, Andrei and Grundmann, Matthias}, title = {Speed Is All You Need: On-Device Acceleration of Large Diffusion Models via GPU-Aware Optimizations}, booktitle = {Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) Workshops}, month = {June}, year = {2023}, pages = {4651-4655} }
Dynamic Inference Acceleration of 3D Point Cloud Deep Neural Networks Using Point Density and Entropy: Gyudo Park,

SooHyeok Kang,

Wencan Cheng,

Jong Hwan Ko; [pdf]
[bibtex]
@InProceedings{Park_2023_CVPR, author = {Park, Gyudo and Kang, SooHyeok and Cheng, Wencan and Ko, Jong Hwan}, title = {Dynamic Inference Acceleration of 3D Point Cloud Deep Neural Networks Using Point Density and Entropy}, booktitle = {Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) Workshops}, month = {June}, year = {2023}, pages = {4725-4729} }
DeepGEMM: Accelerated Ultra Low-Precision Inference on CPU Architectures Using Lookup Tables: Darshan C. Ganji,

Saad Ashfaq,

Ehsan Saboori,

Sudhakar Sah,

Saptarshi Mitra,

MohammadHossein AskariHemmat,

Alexander Hoffman,

Ahmed Hassanien,

Mathieu Léonardon; [pdf] [arXiv]
[bibtex]
@InProceedings{Ganji_2023_CVPR, author = {Ganji, Darshan C. and Ashfaq, Saad and Saboori, Ehsan and Sah, Sudhakar and Mitra, Saptarshi and AskariHemmat, MohammadHossein and Hoffman, Alexander and Hassanien, Ahmed and L\'eonardon, Mathieu}, title = {DeepGEMM: Accelerated Ultra Low-Precision Inference on CPU Architectures Using Lookup Tables}, booktitle = {Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) Workshops}, month = {June}, year = {2023}, pages = {4656-4664} }
Making Models Shallow Again: Jointly Learning To Reduce Non-Linearity and Depth for Latency-Efficient Private Inference: Souvik Kundu,

Yuke Zhang,

Dake Chen,

Peter A. Beerel; [pdf] [arXiv]
[bibtex]
@InProceedings{Kundu_2023_CVPR, author = {Kundu, Souvik and Zhang, Yuke and Chen, Dake and Beerel, Peter A.}, title = {Making Models Shallow Again: Jointly Learning To Reduce Non-Linearity and Depth for Latency-Efficient Private Inference}, booktitle = {Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) Workshops}, month = {June}, year = {2023}, pages = {4685-4689} }
Localized Latent Updates for Fine-Tuning Vision-Language Models: Moritz Ibing,

Isaak Lim,

Leif Kobbelt; [pdf] [arXiv]
[bibtex]
@InProceedings{Ibing_2023_CVPR, author = {Ibing, Moritz and Lim, Isaak and Kobbelt, Leif}, title = {Localized Latent Updates for Fine-Tuning Vision-Language Models}, booktitle = {Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) Workshops}, month = {June}, year = {2023}, pages = {4509-4518} }
Revisiting Class Imbalance for End-to-End Semi-Supervised Object Detection: Purbayan Kar,

Vishal Chudasama,

Naoyuki Onoe,

Pankaj Wasnik; [pdf] [supp]
[bibtex]
@InProceedings{Kar_2023_CVPR, author = {Kar, Purbayan and Chudasama, Vishal and Onoe, Naoyuki and Wasnik, Pankaj}, title = {Revisiting Class Imbalance for End-to-End Semi-Supervised Object Detection}, booktitle = {Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) Workshops}, month = {June}, year = {2023}, pages = {4570-4579} }
Pre-Training Auto-Generated Volumetric Shapes for 3D Medical Image Segmentation: Ryu Tadokoro,

Ryosuke Yamada,

Hirokatsu Kataoka; [pdf]
[bibtex]
@InProceedings{Tadokoro_2023_CVPR, author = {Tadokoro, Ryu and Yamada, Ryosuke and Kataoka, Hirokatsu}, title = {Pre-Training Auto-Generated Volumetric Shapes for 3D Medical Image Segmentation}, booktitle = {Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) Workshops}, month = {June}, year = {2023}, pages = {4740-4745} }
Content-Adaptive Downsampling in Convolutional Neural Networks: Robin Hesse,

Simone Schaub-Meyer,

Stefan Roth; [pdf] [supp] [arXiv]
[bibtex]
@InProceedings{Hesse_2023_CVPR, author = {Hesse, Robin and Schaub-Meyer, Simone and Roth, Stefan}, title = {Content-Adaptive Downsampling in Convolutional Neural Networks}, booktitle = {Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) Workshops}, month = {June}, year = {2023}, pages = {4544-4553} }
Vision Transformers With Mixed-Resolution Tokenization: Tomer Ronen,

Omer Levy,

Avram Golbert; [pdf] [supp] [arXiv]
[bibtex]
@InProceedings{Ronen_2023_CVPR, author = {Ronen, Tomer and Levy, Omer and Golbert, Avram}, title = {Vision Transformers With Mixed-Resolution Tokenization}, booktitle = {Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) Workshops}, month = {June}, year = {2023}, pages = {4613-4622} }
CFDP: Common Frequency Domain Pruning: Samir Khaki,

Weihan Luo; [pdf]
[bibtex]
@InProceedings{Khaki_2023_CVPR, author = {Khaki, Samir and Luo, Weihan}, title = {CFDP: Common Frequency Domain Pruning}, booktitle = {Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) Workshops}, month = {June}, year = {2023}, pages = {4715-4724} }
Similar Class Style Augmentation for Efficient Cross-Domain Few-Shot Learning: Manogna Sreenivas,

Soma Biswas; [pdf]
[bibtex]
@InProceedings{Sreenivas_2023_CVPR, author = {Sreenivas, Manogna and Biswas, Soma}, title = {Similar Class Style Augmentation for Efficient Cross-Domain Few-Shot Learning}, booktitle = {Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) Workshops}, month = {June}, year = {2023}, pages = {4590-4598} }
Quantized Proximal Averaging Networks for Compressed Image Recovery: Nareddy Kartheek Kumar Reddy,

Mani Madhoolika Bulusu,

Praveen Kumar Pokala,

Chandra Sekhar Seelamantula; [pdf] [supp]
[bibtex]
@InProceedings{Reddy_2023_CVPR, author = {Reddy, Nareddy Kartheek Kumar and Bulusu, Mani Madhoolika and Pokala, Praveen Kumar and Seelamantula, Chandra Sekhar}, title = {Quantized Proximal Averaging Networks for Compressed Image Recovery}, booktitle = {Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) Workshops}, month = {June}, year = {2023}, pages = {4633-4643} }
Envisioning a Next Generation Extended Reality Conferencing System With Efficient Photorealistic Human Rendering: Chuanyue Shen,

Letian Zhang,

Zhangsihao Yang,

Masood Mortazavi,

Xiyun Song,

Liang Peng,

Heather Yu; [pdf] [supp]
[bibtex]
@InProceedings{Shen_2023_CVPR, author = {Shen, Chuanyue and Zhang, Letian and Yang, Zhangsihao and Mortazavi, Masood and Song, Xiyun and Peng, Liang and Yu, Heather}, title = {Envisioning a Next Generation Extended Reality Conferencing System With Efficient Photorealistic Human Rendering}, booktitle = {Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) Workshops}, month = {June}, year = {2023}, pages = {4623-4632} }
Phase-Field Models for Lightweight Graph Convolutional Networks: Hichem Sahbi; [pdf]
[bibtex]
@InProceedings{Sahbi_2023_CVPR, author = {Sahbi, Hichem}, title = {Phase-Field Models for Lightweight Graph Convolutional Networks}, booktitle = {Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) Workshops}, month = {June}, year = {2023}, pages = {4644-4650} }
AdaMTL: Adaptive Input-Dependent Inference for Efficient Multi-Task Learning: Marina Neseem,

Ahmed Agiza,

Sherief Reda; [pdf] [arXiv]
[bibtex]
@InProceedings{Neseem_2023_CVPR, author = {Neseem, Marina and Agiza, Ahmed and Reda, Sherief}, title = {AdaMTL: Adaptive Input-Dependent Inference for Efficient Multi-Task Learning}, booktitle = {Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) Workshops}, month = {June}, year = {2023}, pages = {4730-4739} }
MARRS: Modern Backbones Assisted Co-Training for Rapid and Robust Semi-Supervised Domain Adaptation: Saurabh Kumar Jain,

Sukhendu Das; [pdf] [supp]
[bibtex]
@InProceedings{Jain_2023_CVPR, author = {Jain, Saurabh Kumar and Das, Sukhendu}, title = {MARRS: Modern Backbones Assisted Co-Training for Rapid and Robust Semi-Supervised Domain Adaptation}, booktitle = {Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) Workshops}, month = {June}, year = {2023}, pages = {4580-4589} }
DynaShare: Task and Instance Conditioned Parameter Sharing for Multi-Task Learning: Elahe Rahimian,

Golara Javadi,

Frederick Tung,

Gabriel Oliveira; [pdf] [arXiv]
[bibtex]
@InProceedings{Rahimian_2023_CVPR, author = {Rahimian, Elahe and Javadi, Golara and Tung, Frederick and Oliveira, Gabriel}, title = {DynaShare: Task and Instance Conditioned Parameter Sharing for Multi-Task Learning}, booktitle = {Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) Workshops}, month = {June}, year = {2023}, pages = {4535-4543} }
Recursions Are All You Need: Towards Efficient Deep Unfolding Networks: Rawwad Alhejaili,

Motaz Alfarraj,

Hamzah Luqman,

Ali Al-Shaikhi; [pdf] [arXiv]
[bibtex]
@InProceedings{Alhejaili_2023_CVPR, author = {Alhejaili, Rawwad and Alfarraj, Motaz and Luqman, Hamzah and Al-Shaikhi, Ali}, title = {Recursions Are All You Need: Towards Efficient Deep Unfolding Networks}, booktitle = {Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) Workshops}, month = {June}, year = {2023}, pages = {4705-4714} }
BlazeStyleGAN: A Real-Time On-Device StyleGAN: Haolin Jia,

Qifei Wang,

Omer Tov,

Yang Zhao,

Fei Deng,

Lu Wang,

Chuo-Ling Chang,

Tingbo Hou,

Matthias Grundmann; [pdf]
[bibtex]
@InProceedings{Jia_2023_CVPR, author = {Jia, Haolin and Wang, Qifei and Tov, Omer and Zhao, Yang and Deng, Fei and Wang, Lu and Chang, Chuo-Ling and Hou, Tingbo and Grundmann, Matthias}, title = {BlazeStyleGAN: A Real-Time On-Device StyleGAN}, booktitle = {Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) Workshops}, month = {June}, year = {2023}, pages = {4690-4694} }
Rethinking Dilated Convolution for Real-Time Semantic Segmentation: Roland Gao; [pdf] [supp] [arXiv]
[bibtex]
@InProceedings{Gao_2023_CVPR, author = {Gao, Roland}, title = {Rethinking Dilated Convolution for Real-Time Semantic Segmentation}, booktitle = {Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) Workshops}, month = {June}, year = {2023}, pages = {4675-4684} }
MIMMO: Multi-Input Massive Multi-Output Neural Network: Martin Ferianc,

Miguel Rodrigues; [pdf] [supp]
[bibtex]
@InProceedings{Ferianc_2023_CVPR, author = {Ferianc, Martin and Rodrigues, Miguel}, title = {MIMMO: Multi-Input Massive Multi-Output Neural Network}, booktitle = {Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) Workshops}, month = {June}, year = {2023}, pages = {4564-4569} }
Dataset Efficient Training With Model Ensembling: Yeonju Ro,

Cong Xu,

Agnieszka Ciborowska,

Suparna Bhattacharya,

Frankie Li,

Martin Foltin; [pdf]
[bibtex]
@InProceedings{Ro_2023_CVPR, author = {Ro, Yeonju and Xu, Cong and Ciborowska, Agnieszka and Bhattacharya, Suparna and Li, Frankie and Foltin, Martin}, title = {Dataset Efficient Training With Model Ensembling}, booktitle = {Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) Workshops}, month = {June}, year = {2023}, pages = {4700-4704} }