Haoqin Hong's repositories
CCSPNet-Joint
(IJCNN 2024 Oral) This is the joint training model for traffic sign detection and image denoising proposed in our paper titled "CCSPNet-Joint: Efficient Joint Training Method for Traffic Sign Detection Under Extreme Conditions". The dataset CCTSDB-AUG has been already released!
Memoristive-Vision-Retentive-Network
This is the algorithm for our memoristor full-circuit implementation of Vision Retentive Network, a single-layer, single-head Encoder structure is used in the paper.
GKD-RoboMaster-Five-Point-Dataset
This is the five point dataset for yolov5-face-base algorithm used by SWU GKD Team in RoboMaster 2023, all the datas and annotations were made by us.
RoboMasterUnionGKD2023
This is the open source dataset and Yolov5-face-based code used by the GKD team in the RoboMaster 2023 competition
MOLED-Drive
This is the corrected drive for MOLED
rm2023_auto_sentry_ws
RM2023华南农业大学Taurus机器人战队哨兵定位导航算法开源
segmentation_models.pytorch
Segmentation models with pretrained backbones. PyTorch.
simpletransformers
Transformers for Classification, NER, QA, Language Modelling, Language Generation, T5, Multi-Modal, and Conversational AI
torchscale
Foundation Architecture for (M)LLMs
Tpu-Graph-Runtime-Prediction
Update later