Benchmark and Model Zoo

September 3, 2021 ยท View on GitHub

Common settings

  • We use distributed training.

  • All pytorch-style pretrained backbones on ImageNet are from PyTorch model zoo.

  • For fair comparison with other codebases, we report the GPU memory as the maximum value of torch.cuda.max_memory_allocated() for all 8 GPUs. Note that this value is usually less than what nvidia-smi shows.

  • We report the inference time as the total time of network forwarding and post-processing, excluding the data loading time. Results are obtained with the script tools/benchmark.py which computes the average time on 2000 images.

  • Speed benchmark environments

    HardWare

    • 8 NVIDIA Tesla V100 (32G) GPUs
    • Intel(R) Xeon(R) Gold 6148 CPU @ 2.40GHz

    Software environment

    • Python 3.7
    • PyTorch 1.5
    • CUDA 10.1
    • CUDNN 7.6.03
    • NCCL 2.4.08

Baselines of video object detection

DFF

Please refer to DFF for details.

FGFA

Please refer to FGFA for details.

SELSA

Please refer to SELSA for details.

Baselines of multiple object tracking

SORT/DeepSORT

Please refer to SORT/DeepSORT for details.

Tracktor

Please refer to Tracktor for details.

Baselines of single object tracking

SiameseRPN++

Please refer to SiameseRPN++ for details.