passion3394

Is that possible to inference more than 4 lanes with the CULane dataset?

6

is that possible to inference more than 4 lanes with the CULane dataset? My task needs to detect more than 4 lanes, what should I do if I want to...

how to do the unsupervised training?

1

thanks for sharing the code, but the main8Xmulti.py script uses disparity gound truth, my question is : how to do the unsupervised training?

try to implement PSMNet use yours instructions

19

stereo_net = Nets.get_stereo_net(args.modelName, net_args) the above code runs good, but when I start to train on scene_flow, got the following error. I have done some debug methods: (1) check the...

I try to train on scene_flow dataset. the train script is the following: CUDA_VISIBLE_DEVICES=4 python Train.py --trainingSet=/host/nfs/hs/scene_flow/train.list --validationSet=/host/nfs/hs/scene_flow/val.list --output=_output_PSMNet --modelName=PSMNet --imageShape 256 512 --batchSize=1 --numEpochs=10

test result on Apollo depth dataset

12

dear JiaRen, I try my best to convert the apollo depth image to the right disparity, it's successful. Then I try to use the disparity to train a model on...

training loss = nan on custom dataset

10

A slice training process is like the following: Iter 496 training loss = 65.174 , time = 2.39 epoch 41 total training loss = nan Iter 497 training loss =...

why does the testing process need more GPU memory than the training process

7

with the same size of pictures. I could train my own dataset with the PSMNet code, the train process needs about 8 G GPU memory, but failed to test on...

why calculating the input_calibration is so slow?

2

I want to calculate the input_calibration para on 39000 pics, it has run about two days, but no para come out. why?

the first iteration, acc: 94.7188, is it normal?

2

2019-07-08 19:39:57,836 - INFO - Epoch [1][50/39274] lr: 0.00040, eta: 2 days, 23:25:49, time: 1.091, data_time: 0.013, memory: 4931, loss_rpn_cls: 0.6880, loss_rpn_reg: 0.0123, loss_cls: 3.2936, acc: 94.7188, loss_reg: 0.0006, loss:...