Meeting date: 07-06-2023 (9h30)

Summary

Single-task

Object detection

All YOLO versions

Lane detection

https://paperswithcode.com/sota/lane-detection-on-culane

https://github.com/Turoad/lanedet (output isn’t segmentation)

→ (CLRNet) https://arxiv.org/pdf/2305.08366v1.pdf

Enet-SAD (SCNN detector)

https://arxiv.org/pdf/1908.00821.pdf

https://github.com/InhwanBae/ENet-SAD_Pytorch - No pre-trained model

https://github.com/cardwing/Codes-for-Lane-Detection

https://drive.google.com/file/d/1-E0Bws7-v35vOVfqEXDTJdfovUTQ2sf5/view?pli=1

LaneNet

https://arxiv.org/pdf/1807.01726v1.pdf

https://github.com/klintan/pytorch-lanenet

https://github.com/IvanVassi/LaneNet-with-homography - No pre-trained model

https://github.com/Turoad/lanedet - different kinds of detectors and pre-trained models

Semantic segmentation

Enet (multiple classes)

https://arxiv.org/pdf/1606.02147.pdf

https://github.com/gjy3035/enet.pytorch

https://github.com/davidtvs/PyTorch-ENet/ - already implemented

EspNet

https://github.com/sacmehta/ESPNet

EdgeNet (only drivable area (and uses multi-task learning techniques))

https://ieeexplore.ieee.org/document/8951276

Can’t find any implementation

Multi-task

Car detection + Lane detection + Drivable area

YOLOP

YOLOPv2 (previously done)

HybridNet

https://arxiv.org/pdf/2203.09035.pdf

https://github.com/datvuthanh/HybridNets

Object deteciton + Semantic segmentation