Tangshitao / QuadTreeAttention

QuadTree Attention for Vision Transformers (ICLR2022)

Tangshitao/QuadTreeAttention Issues

你好，请问一下，我在参考LoFTR训练megadepth中，该如何设置从LoFTR云盘train-data文件中的cfg_1513_-1_0.2_0.8_0.15_reduced_v2文件的路径
Updated a month ago14
Run QuadTreeAttention on CPU
Closed 2 years ago10
安装时遇到问题
Updated 6 months ago
import issue
Closed 2 years ago2
How to train the model for MegaDepth Dataset
Closed 10 months ago2
Attention visualization
Updated a year ago
验证与测试结果相差较大
Updated a year ago
Reproduce_Train on MegaDepth
Updated a year ago
Killed！
Updated a year ago6
Install Problem
Updated a year ago3
Installation Problem(cuda and pytorch version seem MATCH)
Updated a year ago1
RuntimeError: main thread is not in main loop
Updated a year ago1
RuntimeError: CUDA out of memory.
Updated a year ago1
Hello, would you provide the stereo matching code in the paper?
Updated a year ago1
Hello, would you provide the stereo matching code in the paper?
Updated a year ago2
能提供下立体匹配的代码吗
Updated a year ago
Demo on an image pair request
Updated 2 years ago4
请问能分享loftr-lite的配置文件吗？
Updated 2 years ago5
License?
Updated 2 years ago3
Why training is ~3x slower than Swin?
Updated 2 years ago2
Config questions of notebook and scripts of feature matching
Updated 2 years ago2
Support for CUDA version 11.x
Closed 2 years ago2
error: command '/usr/bin/gcc' failed with exit code 1
Closed 2 years ago5
Image resolution for outdoor feature matching
Closed 2 years ago5
n/a
Closed 2 years ago
What is the role of the depth-wise conv in FFN?
Closed 2 years ago
For FeatureMatching, the QuadTreeAttention-based LoFTR is slower than the original LoFTR
Closed 2 years ago1
Running scripts for reproducing causes Segmentation Fault
Closed 2 years ago1
请问您会更新论文中立体匹配方向的代码吗
Closed 2 years ago2