Skip to content

Latest commit

 

History

History
38 lines (22 loc) · 2 KB

File metadata and controls

38 lines (22 loc) · 2 KB

Benchmark and Model Zoo

Common settings

  • We use distributed training with 2 GPUs by default. For different settings such as transformer backbones, we will illustrate in the benchmark.
  • (TODO) For the consistency across different hardwares, we report the GPU memory as the maximum value of torch.cuda.max_memory_allocated() for all 4 GPUs with torch.backends.cudnn.benchmark=False. Note that this value is usually less than what nvidia-smi shows.
  • (TODO) 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 200 images with torch.backends.cudnn.benchmark=False.
  • (TODO) For input size of 8x+1 (e.g. 769), align_corner=True is adopted as a traditional practice. Otherwise, for input size of 8x (e.g. 512, 1024), align_corner=False is adopted. I think there are potential discrepancies here. Take an instance of Adabins, will the input is not 8x+1, it uses align_corner=True in their offical implementation. The influence to results is not proved. More exps TBD.

Baselines

BTS

Please refer to BTS for details.

Adabins

Please refer to Adabins for details.

DPT

This is a simple implementation. Only model structure is aligned with original paper. More experiments about training settings or loss functions are needed to be done.

Please refer to DPT for details.

SimIPU

Please refer to SimIPU for details.

DepthFormer

Please refer to DepthFormer for details.