Skip to content

Latest commit

 

History

History
9 lines (6 loc) · 1015 Bytes

File metadata and controls

9 lines (6 loc) · 1015 Bytes

Finetuning object detection with TorchVision

Open In Colab

  • Developed a custom object detection and instance segmentation model using Mask R-CNN and Faster R-CNN architectures, built on top of pre-trained CNN backbones such as ResNet-50 and MobileNetV2.
  • Used the PyTorch deep learning framework to train and evaluate the models on a custom dataset of images and annotations, achieving high accuracy and performance metrics.
  • Applied techniques such as transfer learning, fine-tuning, and data augmentation to improve the performance of the models on the target task, and optimized the training process using various hyperparameters and loss functions.
  • Visualized the predictions of the models using various tools and libraries such as Matplotlib, NumPy, and OpenCV, and provided insights and feedback to improve the models based on the observed results.