EfficientViT: Multi-Scale Linear Attention for High-Resolution Dense Prediction [paper]

Efficient vision foundation models for high-resolution generation and perception.

News

• (🔥 New) [2025/01/14] We released DC-AE+USiT models: model, training. Using the default training settings and sampling strategy, DC-AE+USiT-2B achieves 1.72 FID on ImageNet 512x512, surpassing the SOTA diffusion model EDM2-XXL and SOTA auto-regressive image generative models (MAGVIT-v2 and MAR-L).
• (🔥 New) [2024/12/24] diffusers supports DC-AE models. All DC-AE models in diffusers safetensors are released. Usage.
• [2024/10/21] DC-AE and EfficientViT block are used in our latest text-to-image diffusion model SANA! Check the project page for more details.
• [2024/10/15] We released Deep Compression Autoencoder (DC-AE): link!
• [2024/07/10] EfficientViT is used as the backbone in Grounding DINO 1.5 Edge for efficient open-set object detection.
• [2024/07/10] EfficientViT-SAM is used in MedficientSAM, the 1st place model in CVPR 2024 Segment Anything In Medical Images On Laptop Challenge.
• [2024/04/23] We released the training code of EfficientViT-SAM.
• [2024/04/06] EfficientViT-SAM is accepted by eLVM@CVPR'24.
• [2024/03/19] Online demo of EfficientViT-SAM is available: https://evitsam.hanlab.ai/.
• [2024/02/07] We released EfficientViT-SAM, the first accelerated SAM model that matches/outperforms SAM-ViT-H's zero-shot performance, delivering the SOTA performance-efficiency trade-off.
• [2023/11/20] EfficientViT is available in the NVIDIA Jetson Generative AI Lab.
• [2023/09/12] EfficientViT is highlighted by MIT home page and MIT News.
• [2023/07/18] EfficientViT is accepted by ICCV 2023.

Content

Deep Compression Autoencoder for Efficient High-Resolution Diffusion Models [paper] [readme]

Deep Compression Autoencoder (DC-AE) is a new family of high-spatial compression autoencoders with a spatial compression ratio of up to 128 while maintaining reconstruction quality. It accelerates all latent diffusion models regardless of the diffusion model architecture.

Demo

Figure 1: We address the reconstruction accuracy drop of high spatial-compression autoencoders.

Figure 2: DC-AE speeds up latent diffusion models.

Figure 3: DC-AE enables efficient text-to-image generation on the laptop. For more details, please check our text-to-image diffusion model SANA.

• Usage of Deep Compression Autoencoder
• Usage of DC-AE-Diffusion
• Evaluate Deep Compression Autoencoder
• Demo DC-AE-Diffusion Models
• Evaluate DC-AE-Diffusion Models
• Train DC-AE-Diffusion Models
• Reference

EfficientViT-SAM: Accelerated Segment Anything Model Without Accuracy Loss [paper] [online demo] [readme]

EfficientViT-SAM is a new family of accelerated segment anything models by replacing SAM's heavy image encoder with EfficientViT. It delivers a 48.9x measured TensorRT speedup on A100 GPU over SAM-ViT-H without sacrificing accuracy.

• Pretrained EfficientViT-SAM Models
• Usage of EfficientViT-SAM
• Evaluate EfficientViT-SAM
• Visualize EfficientViT-SAM
• Deploy EfficientViT-SAM
• Train EfficientViT-SAM
• Reference

EfficientViT-Classification [paper] [readme]

Efficient image classification models with EfficientViT backbones.

• Pretrained EfficientViT Classification Models
• Usage of EfficientViT Classification Models
• Evaluate EfficientViT Classification Models
• Export EfficientViT Classification Models
• Train EfficientViT Classification Models
• Reference

EfficientViT-Segmentation [paper] [readme]

Efficient semantic segmantation models with EfficientViT backbones.

• Pretrained EfficientViT Segmentation Models
• Usage of EfficientViT Segmentation Models
• Evaluate EfficientViT Segmentation Models
• Visualize EfficientViT Segmentation Models
• Export EfficientViT Segmentation Models
• Reference

EfficientViT-GazeSAM [readme]

Gaze-prompted image segmentation models capable of running in real time with TensorRT on an NVIDIA RTX 4070.

Getting Started

conda create -n efficientvit python=3.10
conda activate efficientvit
pip install -U -r requirements.txt

Third-Party Implementation/Integration

• NVIDIA Jetson Generative AI Lab
• timm: link
• X-AnyLabeling: link
• Grounding DINO 1.5 Edge: link

Contact

Han Cai

Reference

If EfficientViT or EfficientViT-SAM or DC-AE is useful or relevant to your research, please kindly recognize our contributions by citing our paper:

@inproceedings{cai2023efficientvit,
  title={Efficientvit: Lightweight multi-scale attention for high-resolution dense prediction},
  author={Cai, Han and Li, Junyan and Hu, Muyan and Gan, Chuang and Han, Song},
  booktitle={Proceedings of the IEEE/CVF International Conference on Computer Vision},
  pages={17302--17313},
  year={2023}
}

@article{zhang2024efficientvit,
  title={EfficientViT-SAM: Accelerated Segment Anything Model Without Performance Loss},
  author={Zhang, Zhuoyang and Cai, Han and Han, Song},
  journal={arXiv preprint arXiv:2402.05008},
  year={2024}
}

@article{chen2024deep,
  title={Deep Compression Autoencoder for Efficient High-Resolution Diffusion Models},
  author={Chen, Junyu and Cai, Han and Chen, Junsong and Xie, Enze and Yang, Shang and Tang, Haotian and Li, Muyang and Lu, Yao and Han, Song},
  journal={arXiv preprint arXiv:2410.10733},
  year={2024}
}