Event-based Streaming Vision Transformer

Event-based Streaming Vision Transformerfor Challenging Object Detection

Comparison with other state-of-the-art methods

Conda Installation

We train our models underpython=3.8,pytorch=2.2.2,cuda=11.8.

1. Install Pytorch and torchvision. Follow the instruction on https://pytorch.org/get-started/locally/.

conda install -c pytorch pytorch torchvision

2. Install other needed packages

pip install -r requirements.txt

3. For more detailed installation steps, please refer to RTDETR

Demo

1. We have put our model checkpoints here ESVT: Goole Drive
2. We have supplemented and modified the dataset, and the final dataset EMRS Dataset can be downloaded from Baidu Netdisk.
3. Please download weights and organize them as following: weights

  └── ESVT/

      └── ESVT-sbfpn-6xd.pth

4. Run train.py and the results in outputs.