lx-r

update annotation of input size in demo_ncnn/nanodet.h to avoid ambiguity.

What's the purpose of the multipication of n_classes to label in evaluation code?

1

https://github.com/MichaelFan01/STDC-Seg/blob/59ff37fbd693b99972c76fcefe97caa14aeb619f/evaluation.py#L56-L59

nanodet-plus的torchscript模型转rknn报错`E AttributeError: 'numpy.ndarray' object has no attribute 'url'`

4

``` E Catch exception when loading pytorch model: nanodet.torchscript.pt! E Traceback (most recent call last): E File "rknn/base/RKNNlib/converter/convert_pytorch_new.py", line 3985, in rknn.base.RKNNlib.converter.convert_pytorch_new.convert_pytorch.load E File "rknn/base/RKNNlib/converter/convert_pytorch_new.py", line 4156, in rknn.base.RKNNlib.converter.convert_pytorch_new.convert_pytorch.parse_nets E...

第7章,ch7 pose_estimation_2d2d.cpp求得的基础矩阵和本质矩阵E关系没有满足严格的推导公式的相等关系

1

```python F = [[ 4.54443750e-06 1.33385558e-04 -1.79849925e-02] [-1.27565701e-04 2.26679480e-05 -1.41667843e-02] [ 1.81499464e-02 4.14605587e-03 1.00000000e+00]] E = [[ 0.01097677 0.24837205 0.03167429] [-0.20888332 0.02908424 -0.67446588] [ 0.00828678 0.66140416 0.01676524]] K = [[520.9...