Awesome

Official PyTorch implementation of "MobileMamba: Lightweight Multi-Receptive Visual Mamba Network".

Haoyang He^1*, Jiangning Zhang^2*, Yuxuan Cai³, Hongxu Chen¹ Xiaobin Hu²,

Zhenye Gan², Yabiao Wang², Chengjie Wang², Yunsheng Wu², Lei Xie^1†

¹College of Control Science and Engineering, Zhejiang University, ²Youtu Lab, Tencent, ³Huazhong University of Science and Technology

Abstract: Previous research on lightweight models has primarily focused on CNNs and Transformer-based designs. CNNs, with their local receptive fields, struggle to capture long-range dependencies, while Transformers, despite their global modeling capabilities, are limited by quadratic computational complexity in high-resolution scenarios. Recently, state-space models have gained popularity in the visual domain due to their linear computational complexity. Despite their low FLOPs, current lightweight Mamba-based models exhibit suboptimal throughput. In this work, we propose the MobileMamba framework, which balances efficiency and performance. We design a three-stage network to enhance inference speed significantly. At a fine-grained level, we introduce the Multi-Receptive Field Feature Interaction MRFFI module, comprising the Long-Range Wavelet Transform-Enhanced Mamba WTE-Mamba, Efficient Multi-Kernel Depthwise Convolution MK-DeConv, and Eliminate Redundant Identity components. This module integrates multi-receptive field information and enhances high-frequency detail extraction. Additionally, we employ training and testing strategies to further improve performance and efficiency. MobileMamba achieves up to 83.6% on Top-1, surpassing existing state-of-the-art methods which is maximum x21 faster than LocalVim on GPU. Extensive experiments on high-resolution downstream tasks demonstrate that MobileMamba surpasses current efficient models, achieving an optimal balance between speed and accuracy.

Top: Visualization of the Effective Receptive Fields (ERF) for different architectures. Bottom: Performance vs. FLOPs with recent CNN/Transformer/Mamba-based methods.<br>

Accuracy vs. Speed with Mamba-based methods.

Classification results

Image Classification for ImageNet-1K:

Downstream Results

Object Detection and Instant Segmentation Results

Object Detection and Instant Segmentation Performance Based on Mask-RCNN for COCO2017:

Object Detection Performance Based on RetinaNet for COCO2017:

Object Detection Performance Based on SSDLite for COCO2017:

Semantic Segmentation Results

Semantic Segmentation Based on Semantic FPN for ADE20k:

Semantic Segmentation Based on DeepLabv3 for ADE20k:

Semantic Segmentation Based on PSPNet for ADE20k:

All Pretrained Weights and Logs

The model weights and log files for all classification and downstream tasks are available for download via weights.

Classification

Environments

pip3 install torch==2.1.2 torchvision==0.16.2 torchaudio==2.1.2 --index-url https://download.pytorch.org/whl/cu118
pip3 install timm==0.6.5 tensorboardX einops torchprofile fvcore==0.1.5.post20221221 
cd model/lib_mamba/kernels/selective_scan && pip install . && cd ../../../..
git clone https://github.com/NVIDIA/apex && cd apex && pip3 install -v --disable-pip-version-check --no-cache-dir --global-option="--cpp_ext" --global-option="--cuda_ext" ./ (optional)

Prepare ImageNet-1K Dataset

Download and extract ImageNet-1K dataset in the following directory structure:

├── imagenet
    ├── train
        ├── n01440764
            ├── n01440764_10026.JPEG
            ├── ...
        ├── ...
    ├── train.txt (optional)
    ├── val
        ├── n01440764
            ├── ILSVRC2012_val_00000293.JPEG
            ├── ...
        ├── ...
    └── val.txt (optional)

Test

Test with 8 GPUs in one node:

python3 -m torch.distributed.launch --nproc_per_node=8 --nnodes=1 --use_env run.py -c configs/mobilemamba/mobilemamba_t2 -m test model.model_kwargs.checkpoint_path=weights/MobileMamba_T2/mobilemamba_t2.pth

This should give Top-1: 73.638 (Top-5: 91.422)

python3 -m torch.distributed.launch --nproc_per_node=8 --nnodes=1 --use_env run.py -c configs/mobilemamba/mobilemamba_t2s -m test model.model_kwargs.checkpoint_path=weights/MobileMamba_T2s/mobilemamba_t2s.pth

This should give Top-1: 76.934 (Top-5: 93.100)

python3 -m torch.distributed.launch --nproc_per_node=8 --nnodes=1 --use_env run.py -c configs/mobilemamba/mobilemamba_t4 -m test model.model_kwargs.checkpoint_path=weights/MobileMamba_T4/mobilemamba_t4.pth

This should give Top-1: 76.086 (Top-5: 92.772)

python3 -m torch.distributed.launch --nproc_per_node=8 --nnodes=1 --use_env run.py -c configs/mobilemamba/mobilemamba_t4s -m test model.model_kwargs.checkpoint_path=weights/MobileMamba_T4s/mobilemamba_t4s.pth

This should give Top-1: 78.914 (Top-5: 94.160)

python3 -m torch.distributed.launch --nproc_per_node=8 --nnodes=1 --use_env run.py -c configs/mobilemamba/mobilemamba_s6 -m test model.model_kwargs.checkpoint_path=weights/MobileMamba_S6/mobilemamba_s6.pth

This should give Top-1: 78.002 (Top-5: 93.992)

python3 -m torch.distributed.launch --nproc_per_node=8 --nnodes=1 --use_env run.py -c configs/mobilemamba/mobilemamba_s6s -m test model.model_kwargs.checkpoint_path=weights/MobileMamba_S6s/mobilemamba_s6s.pth

This should give Top-1: 80.742 (Top-5: 95.182)

python3 -m torch.distributed.launch --nproc_per_node=8 --nnodes=1 --use_env run.py -c configs/mobilemamba/mobilemamba_b1 -m test model.model_kwargs.checkpoint_path=weights/MobileMamba_B1/mobilemamba_b1.pth

This should give Top-1: 79.948 (Top-5: 94.924)

python3 -m torch.distributed.launch --nproc_per_node=8 --nnodes=1 --use_env run.py -c configs/mobilemamba/mobilemamba_b1s -m test model.model_kwargs.checkpoint_path=weights/MobileMamba_B1s/mobilemamba_b1s.pth

This should give Top-1: 82.234 (Top-5: 95.872)

python3 -m torch.distributed.launch --nproc_per_node=8 --nnodes=1 --use_env run.py -c configs/mobilemamba/mobilemamba_b2 -m test model.model_kwargs.checkpoint_path=weights/MobileMamba_B2/mobilemamba_b2.pth

This should give Top-1: 81.624 (Top-5: 95.890)

python3 -m torch.distributed.launch --nproc_per_node=8 --nnodes=1 --use_env run.py -c configs/mobilemamba/mobilemamba_b2s -m test model.model_kwargs.checkpoint_path=weights/MobileMamba_B2s/mobilemamba_b2s.pth

This should give Top-1: 83.260 (Top-5: 96.438)

python3 -m torch.distributed.launch --nproc_per_node=8 --nnodes=1 --use_env run.py -c configs/mobilemamba/mobilemamba_b4 -m test model.model_kwargs.checkpoint_path=weights/MobileMamba_B4/mobilemamba_b4.pth

This should give Top-1: 82.496 (Top-5: 96.252)

python3 -m torch.distributed.launch --nproc_per_node=8 --nnodes=1 --use_env run.py -c configs/mobilemamba/mobilemamba_b4s -m test model.model_kwargs.checkpoint_path=weights/MobileMamba_B4s/mobilemamba_b4s.pth

This should give Top-1: 83.644 (Top-5: 96.606)

Train

Train with 8 GPUs in one node:

python3 -m torch.distributed.launch --nproc_per_node=8 --nnodes=1 --use_env run.py -c configs/mobilemamba/mobilemamba_t2 -m train

python3 -m torch.distributed.launch --nproc_per_node=8 --nnodes=1 --use_env run.py -c configs/mobilemamba/mobilemamba_t2s -m train

python3 -m torch.distributed.launch --nproc_per_node=8 --nnodes=1 --use_env run.py -c configs/mobilemamba/mobilemamba_t4 -m train

python3 -m torch.distributed.launch --nproc_per_node=8 --nnodes=1 --use_env run.py -c configs/mobilemamba/mobilemamba_t4s -m train

python3 -m torch.distributed.launch --nproc_per_node=8 --nnodes=1 --use_env run.py -c configs/mobilemamba/mobilemamba_s6 -m train

python3 -m torch.distributed.launch --nproc_per_node=8 --nnodes=1 --use_env run.py -c configs/mobilemamba/mobilemamba_s6s -m train

python3 -m torch.distributed.launch --nproc_per_node=8 --nnodes=1 --use_env run.py -c configs/mobilemamba/mobilemamba_b1 -m train

python3 -m torch.distributed.launch --nproc_per_node=8 --nnodes=1 --use_env run.py -c configs/mobilemamba/mobilemamba_b1s -m train

python3 -m torch.distributed.launch --nproc_per_node=8 --nnodes=1 --use_env run.py -c configs/mobilemamba/mobilemamba_b2 -m train

python3 -m torch.distributed.launch --nproc_per_node=8 --nnodes=1 --use_env run.py -c configs/mobilemamba/mobilemamba_b2s -m train

python3 -m torch.distributed.launch --nproc_per_node=8 --nnodes=1 --use_env run.py -c configs/mobilemamba/mobilemamba_b4 -m train

python3 -m torch.distributed.launch --nproc_per_node=8 --nnodes=1 --use_env run.py -c configs/mobilemamba/mobilemamba_b4s -m train

Down-Stream Tasks

Environments

pip3 install terminaltables pycocotools prettytable xtcocotools
pip3 install mmpretrain==1.2.0 mmdet==3.3.0 mmsegmentation==1.2.2
pip3 install mmcv==2.1.0 -f https://download.openmmlab.com/mmcv/dist/cu118/torch2.1/index.html
cd det/backbones/lib_mamba/kernels/selective_scan && pip install . && cd ../../../..

Prepare COCO and ADE20k Dataset

Download and extract COCO2017 and ADE20k dataset in the following directory structure:

downstream
├── det
├──── data
│   ├──── coco
│   │   ├──── annotations
│   │   ├──── train2017
│   │   ├──── val2017
│   │   ├──── test2017
├── seg
├──── data
│   ├──── ade
│   │   ├──── ADEChallengeData2016
│   │   ├──────── annotations
│   │   ├──────── images

Object Detection

Train:

CUDA_VISIBLE_DEVICES=0,1,2,3 ./tools/dist_train.sh configs/mask_rcnn/mask-rcnn_mobilemamba_b1_fpn_1x_coco.py 4

Test:

CUDA_VISIBLE_DEVICES=0,1,2,3 ./tools/dist_test.sh configs/mask_rcnn/mask-rcnn_mobilemamba_b1_fpn_1x_coco.py ../../weights/downstream/det/maskrcnn.pth 4

Train:

CUDA_VISIBLE_DEVICES=0,1,2,3 ./tools/dist_train.sh configs/retinanet/retinanet_mobilemamba_b1_fpn_1x_coco.py 4

Test:

CUDA_VISIBLE_DEVICES=0,1,2,3 ./tools/dist_test.sh configs/retinanet/retinanet_mobilemamba_b1_fpn_1x_coco.py ../../weights/downstream/det/retinanet.pth 4

Train with 320 x 320 resolution:

./tools/dist_train.sh configs/ssd/ssdlite_mobilemamba_b1_8gpu_2lr_coco.py 8

Test with 320 x 320 resolution:

./tools/dist_test.sh configs/ssd/ssdlite_mobilemamba_b1_8gpu_2lr_coco.py ../../weights/downstream/det/ssdlite.pth 8

Train with 512 x 512 resolution:

./tools/dist_train.sh configs/ssd/ssdlite_mobilemamba_b1_8gpu_2lr_512_coco.py 8

Test with 512 x 512 resolution:

./tools/dist_test.sh configs/ssd/ssdlite_mobilemamba_b1_8gpu_2lr_512_coco.py ../../weights/downstream/det/ssdlite_512.pth 8

Semantic Segmentation

Train:

CUDA_VISIBLE_DEVICES=0,1,2,3 ./tools/dist_train.sh configs/deeplabv3/deeplabv3_mobilemamba_b4-80k_ade20k-512x512.py 4

Test:

CUDA_VISIBLE_DEVICES=0,1,2,3 ./tools/dist_test.sh configs/deeplabv3/deeplabv3_mobilemamba_b4-80k_ade20k-512x512.py ../../weights/downstream/seg/deeplabv3.pth 4

Train:

CUDA_VISIBLE_DEVICES=0,1,2,3 ./tools/dist_train.sh configs/sem_fpn/fpn_mobilemamba_b4-160k_ade20k-512x512.py 4

Test:

CUDA_VISIBLE_DEVICES=0,1,2,3 ./tools/dist_test.sh configs/sem_fpn/fpn_mobilemamba_b4-160k_ade20k-512x512.py ../../weights/downstream/seg/fpn.pth 4

Train:

CUDA_VISIBLE_DEVICES=0,1,2,3 ./tools/dist_train.sh configs/pspnet/pspnet_mobilemamba_b4-80k_ade20k-512x512.py 4

Test:

CUDA_VISIBLE_DEVICES=0,1,2,3 ./tools/dist_test.sh configs/pspnet/pspnet_mobilemamba_b4-80k_ade20k-512x512.py ../../weights/downstream/seg/pspnet.pth 4

Citation

If our work is helpful for your research, please consider citing:

@article{mobilemamba,
  title={MobileMamba: Lightweight Multi-Receptive Visual Mamba Network},
  author={Haoyang He and Jiangning Zhang and Yuxuan Cai and Hongxu Chen and Xiaobin Hu and Zhenye Gan and Yabiao Wang and Chengjie Wang and Yunsheng Wu and Lei Xie},
  journal={arXiv preprint arXiv:2411.15941},
  year={2024}
}

Acknowledgements

We thank but not limited to following repositories for providing assistance for our research:

• EMO
• EfficientViT
• VMamba
• TIMM
• MMDetection
• MMSegmentation