{"id":13856919,"url":"https://github.com/linksense/LightNetPlusPlus","last_synced_at":"2025-07-13T19:33:40.129Z","repository":{"id":109420918,"uuid":"172831995","full_name":"linksense/LightNetPlusPlus","owner":"linksense","description":"LightNet++: Boosted Light-weighted Networks for Real-time Semantic Segmentation","archived":false,"fork":false,"pushed_at":"2020-03-24T10:42:58.000Z","size":67720,"stargazers_count":238,"open_issues_count":5,"forks_count":54,"subscribers_count":15,"default_branch":"master","last_synced_at":"2024-05-15T04:39:38.687Z","etag":null,"topics":["apex","aspp","attention-mechanism","bifpn","cityscapes","deepdrive","deeplab-v3-plus","dense-aspp","efficientnet","inplaceabn","light-weighted-network","lightnet","mixnet","mobilenetv2","mobilenetv2plus","pytorch","semantic-segmentation","shufflenetv2","shufflenetv2plus","unsharp-masking"],"latest_commit_sha":null,"homepage":"https://ansleliu.github.io/LightNet.html","language":"Python","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":"mit","status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/linksense.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":null,"funding":null,"license":"LICENSE","code_of_conduct":null,"threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":null,"support":null,"governance":null}},"created_at":"2019-02-27T02:59:04.000Z","updated_at":"2024-04-05T05:59:18.000Z","dependencies_parsed_at":"2023-05-20T10:15:48.562Z","dependency_job_id":null,"html_url":"https://github.com/linksense/LightNetPlusPlus","commit_stats":null,"previous_names":[],"tags_count":0,"template":false,"template_full_name":null,"repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/linksense%2FLightNetPlusPlus","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/linksense%2FLightNetPlusPlus/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/linksense%2FLightNetPlusPlus/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/linksense%2FLightNetPlusPlus/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/linksense","download_url":"https://codeload.github.com/linksense/LightNetPlusPlus/tar.gz/refs/heads/master","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":213988286,"owners_count":15666956,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2022-07-04T15:15:14.044Z","host_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub","repositories_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories","repository_names_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repository_names","owners_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners"}},"keywords":["apex","aspp","attention-mechanism","bifpn","cityscapes","deepdrive","deeplab-v3-plus","dense-aspp","efficientnet","inplaceabn","light-weighted-network","lightnet","mixnet","mobilenetv2","mobilenetv2plus","pytorch","semantic-segmentation","shufflenetv2","shufflenetv2plus","unsharp-masking"],"created_at":"2024-08-05T03:01:18.835Z","updated_at":"2025-10-14T12:10:09.886Z","avatar_url":"https://github.com/linksense.png","language":"Python","funding_links":[],"categories":["Python"],"sub_categories":[],"readme":"# LightNet++\n\n## !!!New Repo.!!! ⇒ **[EfficientNet.PyTorch: Concise, Modular, Human-friendly PyTorch implementation of EfficientNet with Pre-trained Weights](https://github.com/ansleliu/EfficientNet.PyTorch)**\n## !!!New Repo.!!! ⇒ **[MixNet-Pytorch: Concise, Modular, Human-friendly PyTorch implementation of MixNet with Pre-trained Weights](https://github.com/ansleliu/MixNet-Pytorch)**\n\nThis repository contains the code (PyTorch-1.0+, **W.I.P.**) for: \"**LightNet++: Boosted Light-weighted Networks for Real-time Semantic Segmentation**\" by Huijun Liu.  \n**LightNet++** is an advanced version of **[LightNet](https://github.com/ansleliu/LightNet)**,  which purpose to get more concise model design, \nsmaller models, and better performance.\n\n- **MobileNetV2Plus**: Modified MobileNetV2 (backbone)\u003csup\u003e[[1,8]](#references)\u003c/sup\u003e + DSASPPInPlaceABNBlock\u003csup\u003e[[2,3]](#references)\u003c/sup\u003e + \nParallel Bottleneck Channel-Spatial Attention Block (PBCSABlock)\u003csup\u003e[[6]](#references)\u003c/sup\u003e + UnSharp Masking (USM) + Encoder-Decoder Arch.\u003csup\u003e[[3]](#references)\u003c/sup\u003e + \nInplaceABN\u003csup\u003e[[4]](#references)\u003c/sup\u003e.\n\n- **ShuffleNetV2Plus**: Modified ShuffleNetV2 (backbone)\u003csup\u003e[[1,8]](#references)\u003c/sup\u003e + DSASPPInPlaceABNBlock\u003csup\u003e[[2,3]](#references)\u003c/sup\u003e + \nParallel Bottleneck Channel-Spatial Attention Block (PBCSABlock)\u003csup\u003e[[6]](#references)\u003c/sup\u003e+ UnSharp Masking (USM)  + Encoder-Decoder Arch.\u003csup\u003e[[3]](#references)\u003c/sup\u003e + \nInplaceABN\u003csup\u003e[[4]](#references)\u003c/sup\u003e.\n\n- [**MixSeg-MixBiFPN**](https://github.com/ansleliu/LightNetPlusPlus/blob/master/models/mixnetseg.py): Modified MixNet (backbone)\u003csup\u003e[[1,8]](#references)\u003c/sup\u003e + MixBiFPNBlock\u003csup\u003e[[2,3]](#references)\u003c/sup\u003e + Encoder-Decoder Arch.\u003csup\u003e[[3]](#references)\u003c/sup\u003e\n \nMore about **USM(Unsharp Mask)-Operator Block** see Repo: [**SharpPeleeNet**](https://github.com/ansleliu/SharpPeleeNet)\n\n## Dependencies\n\n- [Python3.6](https://www.python.org/downloads/)  \n- [PyTorch(1.0.1+)](http://pytorch.org)  \n- [inplace_abn](https://github.com/mapillary/inplace_abn)  \n- [apex](https://github.com/NVIDIA/apex): Tools for easy mixed precision and distributed training in Pytorch  \n- [tensorboard](https://www.tensorflow.org/programmers_guide/summaries_and_tensorboard)  \n- [tensorboardX](https://github.com/lanpa/tensorboard-pytorch)  \n- [tqdm](https://github.com/tqdm/tqdm)  \n\n### Datasets for Autonomous Driving\n- [Cityscapes](https://www.cityscapes-dataset.com/)  \n- [Mapillary Vistas Dataset](https://www.mapillary.com/dataset/vistas)  \n- [Berkeley Deep Drive (BDD100K)](https://bdd-data.berkeley.edu/)  \n- [ApolloScape](http://apolloscape.auto/index.html#)  \n\n\n## Results\n\n### Results on Cityscapes (Pixel-level/Semantic Segmentation)\n\n| Model | mIoU (S.S* Mixed Precision) |Model Weight|\n|---|---|---|\n|**MobileNetV2Plus X1.0**|[71.5314 (**WIP**)](https://github.com/ansleliu/LightNetPlusPlus/blob/master/checkpoint/MobileNetv2Plus.csv)|[cityscapes_mobilenetv2plus_x1.0.pkl (14.3 MB)](https://github.com/ansleliu/LightNetPlusPlus/blob/master/checkpoint/cityscapes_mobilenetv2plus_x1.0.pkl)|\n|**ShuffleNetV2Plus X1.0**|[69.0885-72.5255 (**WIP**)](https://github.com/ansleliu/LightNetPlusPlus/blob/master/checkpoint/ShuffleNetV2PlusX1.0.csv)|[cityscapes_shufflenetv2plus_x1.0.pkl (8.59 MB)](https://github.com/ansleliu/LightNetPlusPlus/blob/master/checkpoint/cityscapes_shufflenetv2plus_x1.0.pkl)|\n|**MixSeg+MixBiFPN  ArchS**|[72.2321 (**WIP**)](https://github.com/ansleliu/LightNetPlusPlus/blob/master/checkpoint/MixSegMixBiFPN_ArchS.csv)|[cityscapes_mixseg_archs_mixbifpn.pkl (16.4 MB)](https://github.com/ansleliu/LightNetPlusPlus/blob/master/checkpoint/cityscapes_mixseg_archs_mixbifpn.pkl)|\n\n* S.S.: Single Scale (1024x2048)\n\n## Feature Visualization\n\n\u003cp align=\"center\"\u003e\n\u003cimg src=\"https://github.com/ansleliu/LightNetPlusPlus/blob/master/netviz/feat_viz/Figure_1.png\" /\u003e\n\u003cimg src=\"https://github.com/ansleliu/LightNetPlusPlus/blob/master/netviz/feat_viz/Figure_3.png\" /\u003e\n\u003cimg src=\"https://github.com/ansleliu/LightNetPlusPlus/blob/master/netviz/feat_viz/Figure_5.png\" /\u003e\n\u003cimg src=\"https://github.com/ansleliu/LightNetPlusPlus/blob/master/netviz/feat_viz/Figure_8.png\" /\u003e\n\u003c/p\u003e\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Flinksense%2FLightNetPlusPlus","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Flinksense%2FLightNetPlusPlus","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Flinksense%2FLightNetPlusPlus/lists"}