Skip to content

Releases: Westlake-AI/MogaNet

MogaNet-Pose-Estimation-Weights

13 Feb 07:57
@Lupin1998 Lupin1998
moganet-pose-weights
c29129a
Compare
Choose a tag to compare
MogaNet-Pose-Estimation-Weights Latest
Latest

A collection of log.json and model.pth for 2D human pose estimation experiments of MogaNet on COCO (download). You can also download all released files from Baidu Cloud (z8mf) at MogaNet/COCO_Pose.

  • We perform top-down pose estimation experiments based on with ImageNet-1K pre-trained MogaNet variants fine-tuning 210 epochs in MogaNet/pose_estimation. We also provide results of popular architectures (Swin, ConvNeXt, and Uniformer) for comparison.

MogaNet + Top-Down

Backbone Pretrain Input Size Params FLOPs Epoch AP AR Config Download
MogaNet-XT ImageNet-1K 256x192 5.6M 1.84G 210 72.1 77.7 config log / model
MogaNet-XT ImageNet-1K 384x288 5.6M 4.15G 210 74.7 79.9 config log / model
MogaNet-T ImageNet-1K 256x192 8.1M 2.15G 210 73.2 78.8 config log / model
MogaNet-T ImageNet-1K 384x288 8.1M 4.85G 210 75.7 80.9 config log / model
MogaNet-S ImageNet-1K 256x192 29.0M 5.99G 210 74.8 80.1 config log / model
MogaNet-S ImageNet-1K 384x288 29.0M 13.48G 210 76.4 81.4 config log / model
MogaNet-B ImageNet-1K 256x192 47.4M 10.85G 210 75.3 80.7 config log / model
MogaNet-B ImageNet-1K 384x288 47.4M 24.42G 210 77.3 82.2 config log / model

MetaFormers + Top-Down

Backbone Input Size Params FLOPs AP AP50 AP75 AR ARM ARL Config Download
Swin-T 256x192 32.8M 6.1G 72.4 90.1 80.6 78.2 74.0 84.3 config model | log
Swin-B 256x192 93.0M 18.6G 73.7 90.4 82.0 79.8 74.9 85.7 config model | log
Swin-B 384x288 93.0M 40.1G 75.9 91.0 83.2 78.8 76.5 87.5 config model | log
Swin-L 256x192 203.4M 40.3G 74.3 90.6 82.1 79.8 75.5 86.2 config model | log
Swin-L 384x288 203.4M 86.9G 76.3 91.2 83.0 81.4 77.0 87.9 config model | log
ConvNeXt-T 256x192 33.0M 5.5G 73.2 90.0 80.9 78.8 74.5 85.1 config log | model
ConvNeXt-T 384x288 33.0M 12.5G 75.3 90.4 82.1 80.5 76.1 86.8 config log | model
ConvNeXt-S 256x192 54.7M 9.7G 73.7 90.3 81.9 79.3 75.0 85.5 config log | model
ConvNeXt-S 384x288 54.7M 21.8G 75.8 90.7 83.1 81.0 76.8 87.1 config log | model
UniFormer-S 256x192 25.2M 4.7G 74.0 90.3 82.2 79.5 66.8 76.7 config log | model
UniFormer-S 384x288 25.2M 11.1G 75.9 90.6 83.4 81.4 68.6 79.0 config log | [model](https://github.com/Westlake-AI/MogaNet/releases/down...
Read more
Assets 34

MogaNet-ADE20K-Segmentation-Weights

09 Feb 15:18
@Lupin1998 Lupin1998
moganet-seg-weights
5a72c32
Compare
Choose a tag to compare
MogaNet-ADE20K-Segmentation-Weights

A collection of log.json and model.pth for semantic segmentation experiments of MogaNet on ADE20K (download). You can also download all released files from Baidu Cloud (z8mf) at MogaNet/ADE20K_Segmentation.

  • We perform semantic segmentation experiments based on Semantic FPN with ImageNet-1K pre-trained MogaNet variants fine-tuning 80K iterations in MogaNet/segmentation.
  • We perform semantic segmentation experiments based on UperNet with ImageNet-1K pre-trained MogaNet variants fine-tuning 160K iterations in MogaNet/segmentation.

MogaNet + Semantic FPN

Method Backbone Pretrain Params FLOPs Iters mIoU mAcc Config Download
Semantic FPN MogaNet-XT ImageNet-1K 6.9M 101.4G 80K 40.3 52.4 config log / model
Semantic FPN MogaNet-T ImageNet-1K 9.1M 107.8G 80K 43.1 55.4 config log / model
Semantic FPN MogaNet-S ImageNet-1K 29.1M 189.7G 80K 47.7 59.8 config log / model
Semantic FPN MogaNet-B ImageNet-1K 47.5M 293.6G 80K 49.3 61.6 config log / model
Semantic FPN MogaNet-L ImageNet-1K 86.2M 418.7G 80K 50.2 63.0 config log / model

MogaNet + UperNet

Method Backbone Pretrain Params FLOPs Iters mIoU mAcc Config Download
UperNet MogaNet-XT ImageNet-1K 30.4M 855.7G 160K 42.2 55.1 config log / model
UperNet MogaNet-T ImageNet-1K 33.1M 862.4G 160K 43.7 57.1 config log / model
UperNet MogaNet-S ImageNet-1K 55.3M 946.4G 160K 49.2 61.6 config log / model
UperNet MogaNet-B ImageNet-1K 73.7M 1050.4G 160K 50.1 63.4 config log / model
UperNet MogaNet-L ImageNet-1K 113.2M 1176.1G 160K 50.9 63.5 config log / model
Assets 22

MogaNet-COCO-Detection-Weights

09 Feb 14:41
@Lupin1998 Lupin1998
moganet-det-weights
23d17b6
Compare
Choose a tag to compare
MogaNet-COCO-Detection-Weights

A collection of log.json and model.pth for object detection and instance segmentation experiments of MogaNet on COCO2017 (download). You can also download all files from Baidu Cloud (z8mf) at MogaNet/COCO_Detection.

  • We preform object detection experiments based on RetinaNet with ImageNet-1K pre-trained MogaNet variants for 1x training setting in MogaNet/detection.
  • We perform detection and instance segmentation experiments based on Mask R-CNN and Cascade Mask R-CNN with ImageNet-1K pre-trained MogaNet variants for 1x or MS 3x training settings in MogaNet/detection.

MogaNet + RetinaNet

Method Backbone Pretrain Params FLOPs Lr schd box mAP Config Download
RetinaNet MogaNet-XT ImageNet-1K 12.1M 167.2G 1x 39.7 config log / model
RetinaNet MogaNet-T ImageNet-1K 14.4M 173.4G 1x 41.4 config log / model
RetinaNet MogaNet-S ImageNet-1K 35.1M 253.0G 1x 45.8 config log / model
RetinaNet MogaNet-B ImageNet-1K 53.5M 354.5G 1x 47.7 config log / model
RetinaNet MogaNet-L ImageNet-1K 92.4M 476.8G 1x 48.7 config log / model

MogaNet + Mask R-CNN

Method Backbone Pretrain Params FLOPs Lr schd box mAP mask mAP Config Download
Mask R-CNN MogaNet-XT ImageNet-1K 22.8M 185.4G 1x 40.7 37.6 config log / model
Mask R-CNN MogaNet-T ImageNet-1K 25.0M 191.7G 1x 42.6 39.1 config log / model
Mask R-CNN MogaNet-S ImageNet-1K 45.0M 271.6G 1x 46.6 42.2 config log / model
Mask R-CNN MogaNet-B ImageNet-1K 63.4M 373.1G 1x 49.0 43.8 config log / model
Mask R-CNN MogaNet-L ImageNet-1K 102.1M 495.3G 1x 49.4 44.2 config log / model
Mask R-CNN MogaNet-T ImageNet-1K 25.0M 191.7G MS 3x 45.3 40.7 config log / model
Mask R-CNN MogaNet-S ImageNet-1K 45.0M 271.6G MS 3x 48.5 43.1 config log / model
Mask R-CNN MogaNet-B ImageNet-1K 63.4M 373.1G MS 3x 50.3 44.4 config log / model
Mask R-CNN MogaNet-L ImageNet-1K 63.4M 373.1G MS 3x 50.6 44.6 config log / model

MogaNet + Cascade Mask R-CNN

Method Backbone Pretrain Params FLOPs Lr schd box mAP mask mAP Config Download
Cascade Mask R-CNN MogaNet-S ImageNet-1K 77.9M 405.4G MS 3x 51.4 44.9 config log / model
Cascade Mask R-CNN MogaNet-S ImageNet-1K 82.8M 750.2G GIOU+MS 3x 51.7 45.1 config log / model
Cascade Mask R-CNN MogaNet-B ImageNet-1K 101.2M 851.6G GIOU+MS 3x 52.6 46.0 config log / model
Cascade Mask R-CNN MogaNet-L ImageNet-1K 139.9M 973.8G GIOU+MS 3x 53.3 46.1 config -
Assets 36

MogaNet-ImageNet-Weights

13 Jan 21:41
@Lupin1998 Lupin1998
moganet-in1k-weights
c0daabf
Compare
Choose a tag to compare
MogaNet-ImageNet-Weights

A collection of args.yaml, summary.csv, and model.pth.tar for image classification experiments of MogaNet on ImageNet-1K (download). You can download all files from Baidu Cloud: MogaNet (z8mf) at MogaNet/Classification_MogaNet.

  • We reproduce the results of MogaNet for 300-epoch training according to the DeiT setting on ImageNet-1K in TRAINING.md. Refer to OpenMixup for more image classification results.
  • The best top-1 accuracy of image classification of 3 trials is reported for all experiments. Note that we report the classification accuracy of EMA weights for MogaNet-S, MogaNet-B, and MogaNet-L (please evaluate their EMA models).
  • To evaluate the pre-trained weights, use validate.py with scripts for the classification performance.

Image Classification on ImageNet-1K

Model resolution Params Flops Top-1 (%) Top-5 (%) Config Download
MogaNet-XT 224x224 2.97M 0.80G 76.5 93.4 args / config / script model / log
MogaNet-T 224x224 5.20M 1.10G 79.0 94.6 args / config / script model / log
MogaNet-T 256x256 5.20M 1.44G 79.6 94.9 args / config / script model / log
MogaNet-S 224x224 25.3M 4.97G 83.4 96.9 args / config / script model / log
MogaNet-B 224x224 43.9M 9.93G 84.3 97.0 args / config / script model / log
MogaNet-L 224x224 82.5M 15.9G 84.7 97.1 args / config / script model / log
MogaNet-XL 224x224 180.8M 34.5G 85.1 97.4 args / config / script model / log
Assets 22