⭐️ From Baidu.Inc
maybeshewill-cv / mnn-lanenet Goto Github PK
View Code? Open in Web Editor NEWLane detection model for mobile device via MNN project
Lane detection model for mobile device via MNN project
The problem has been solved.
When I make build, I got the following issue:
lanenet_model.cpp:427:10: error: ‘iota’ is not a member of ‘std’
std::iota(indexes.begin(), indexes.end(), 0);
^~~~
My g++ version: g++ (Ubuntu 7.5.0-3ubuntu1~18.04) 7.5.0
CMakefile has set CPP_11 flag on
Do you know how to fix this?
Thanks
你好,首先感谢你的工作。按照你的说明,我成功在linux上跑通了模型,当我想移植到安卓上发现始终无法获得最终结果,在调试的过程中发现以下情况:
在linux上运行 _m_lanenet_model->runSessionWithCallBack(_m_lanenet_session, beforeCallBack, AfterCallBack, true);
可以获取到所有op的name,
LaneNet_1/bisenetv2_frontend/detail_branch/stage_1/conv_block_1_repeat_1/3x3_conv/conv/Conv2D LaneNet_1/bisenetv2_frontend/detail_branch/stage_1/conv_block_2_repeat_1/3x3_conv/conv/Conv2D LaneNet_1/bisenetv2_frontend/detail_branch/stage_2/conv_block_1_repeat_1/3x3_conv/conv/Conv2D LaneNet_1/bisenetv2_frontend/detail_branch/stage_2/conv_block_2_repeat_1/3x3_conv/conv/Conv2D LaneNet_1/bisenetv2_frontend/detail_branch/stage_2/conv_block_2_repeat_2/3x3_conv/conv/Conv2D LaneNet_1/bisenetv2_frontend/detail_branch/stage_3/conv_block_1_repeat_1/3x3_conv/conv/Conv2D LaneNet_1/bisenetv2_frontend/detail_branch/stage_3/conv_block_2_repeat_1/3x3_conv/conv/Conv2D LaneNet_1/bisenetv2_frontend/detail_branch/stage_3/conv_block_2_repeat_2/3x3_conv/conv/Conv2D LaneNet_1/bisenetv2_frontend/aggregation_branch/guided_aggregation_block/detail_branch/3x3_dw_conv_block/depthwise_conv_output LaneNet_1/bisenetv2_frontend/aggregation_branch/guided_aggregation_block/detail_branch/1x1_conv_block/Conv2D LaneNet_1/bisenetv2_frontend/semantic_branch/stage_1/se_block_1_repeat_1/stem_block/conv_block_1/conv/Conv2D LaneNet_1/bisenetv2_frontend/semantic_branch/stage_1/se_block_1_repeat_1/stem_block/downsample_branch_left/1x1_conv_block/conv/Conv2D LaneNet_1/bisenetv2_frontend/semantic_branch/stage_1/se_block_1_repeat_1/stem_block/downsample_branch_left/3x3_conv_block/conv/Conv2D LaneNet_1/bisenetv2_frontend/semantic_branch/stage_1/se_block_1_repeat_1/stem_block/downsample_branch_right/maxpooling_block LaneNet_1/bisenetv2_frontend/semantic_branch/stage_1/se_block_1_repeat_1/stem_block/concate_features LaneNet_1/bisenetv2_frontend/semantic_branch/stage_1/se_block_1_repeat_1/stem_block/final_conv_block/conv/Conv2D LaneNet_1/bisenetv2_frontend/semantic_branch/stage_3/ge_block_1_repeat_1/gather_expansion_block/stride_equal_two_module/input_project_dw_conv_block/depthwise_conv_output LaneNet_1/bisenetv2_frontend/semantic_branch/stage_3/ge_block_1_repeat_1/gather_expansion_block/stride_equal_two_module/input_project_1x1_conv_block/conv/Conv2D LaneNet_1/bisenetv2_frontend/semantic_branch/stage_3/ge_block_1_repeat_1/gather_expansion_block/stride_equal_two_module/3x3_conv_block/conv/Conv2D LaneNet_1/bisenetv2_frontend/semantic_branch/stage_3/ge_block_1_repeat_1/gather_expansion_block/stride_equal_two_module/depthwise_conv_block_1/depthwise_conv_output LaneNet_1/bisenetv2_frontend/semantic_branch/stage_3/ge_block_1_repeat_1/gather_expansion_block/stride_equal_two_module/depthwise_conv_block_2/depthwise_conv_output LaneNet_1/bisenetv2_frontend/semantic_branch/stage_3/ge_block_1_repeat_1/gather_expansion_block/stride_equal_two_module/1x1_conv_block/conv/Conv2D LaneNet_1/bisenetv2_frontend/semantic_branch/stage_3/ge_block_1_repeat_1/gather_expansion_block/stride_equal_two_module/fused_features LaneNet_1/bisenetv2_frontend/semantic_branch/stage_3/ge_block_1_repeat_1/gather_expansion_block/stride_equal_two_module/ge_output LaneNet_1/bisenetv2_frontend/semantic_branch/stage_3/ge_block_2_repeat_1/gather_expansion_block/stride_equal_one_module/3x3_conv_block/conv/Conv2D LaneNet_1/bisenetv2_frontend/semantic_branch/stage_3/ge_block_2_repeat_1/gather_expansion_block/stride_equal_one_module/depthwise_conv_block/depthwise_conv_output LaneNet_1/bisenetv2_frontend/semantic_branch/stage_3/ge_block_2_repeat_1/gather_expansion_block/stride_equal_one_module/1x1_conv_block/conv/Conv2D LaneNet_1/bisenetv2_frontend/semantic_branch/stage_3/ge_block_1_repeat_1/gather_expansion_block/stride_equal_two_module/ge_output___tr4LaneNet_1/bisenetv2_frontend/semantic_branch/stage_3/ge_block_2_repeat_1/gather_expansion_block/stride_equal_one_module/fused_features LaneNet_1/bisenetv2_frontend/semantic_branch/stage_3/ge_block_2_repeat_1/gather_expansion_block/stride_equal_one_module/1x1_conv_block/bn/FusedBatchNorm___tr4LaneNet_1/bisenetv2_frontend/semantic_branch/stage_3/ge_block_2_repeat_1/gather_expansion_block/stride_equal_one_module/fused_features LaneNet_1/bisenetv2_frontend/semantic_branch/stage_3/ge_block_2_repeat_1/gather_expansion_block/stride_equal_one_module/fused_features LaneNet_1/bisenetv2_frontend/semantic_branch/stage_3/ge_block_2_repeat_1/gather_expansion_block/stride_equal_one_module/ge_output LaneNet_1/bisenetv2_frontend/semantic_branch/stage_3/ge_block_2_repeat_1/gather_expansion_block/stride_equal_one_module/ge_output___tr4LaneNet_1/bisenetv2_frontend/semantic_branch/stage_4/ge_block_1_repeat_1/gather_expansion_block/stride_equal_two_module/input_project_dw_conv_block/depthwise_conv_output LaneNet_1/bisenetv2_frontend/semantic_branch/stage_4/ge_block_1_repeat_1/gather_expansion_block/stride_equal_two_module/input_project_dw_conv_block/depthwise_conv_output LaneNet_1/bisenetv2_frontend/semantic_branch/stage_4/ge_block_1_repeat_1/gather_expansion_block/stride_equal_two_module/input_project_1x1_conv_block/conv/Conv2D LaneNet_1/bisenetv2_frontend/semantic_branch/stage_3/ge_block_2_repeat_1/gather_expansion_block/stride_equal_one_module/ge_output___tr4LaneNet_1/bisenetv2_frontend/semantic_branch/stage_4/ge_block_1_repeat_1/gather_expansion_block/stride_equal_two_module/3x3_conv_block/conv/Conv2D LaneNet_1/bisenetv2_frontend/semantic_branch/stage_4/ge_block_1_repeat_1/gather_expansion_block/stride_equal_two_module/3x3_conv_block/conv/Conv2D LaneNet_1/bisenetv2_frontend/semantic_branch/stage_4/ge_block_1_repeat_1/gather_expansion_block/stride_equal_two_module/depthwise_conv_block_1/depthwise_conv_output LaneNet_1/bisenetv2_frontend/semantic_branch/stage_4/ge_block_1_repeat_1/gather_expansion_block/stride_equal_two_module/depthwise_conv_block_2/depthwise_conv_output LaneNet_1/bisenetv2_frontend/semantic_branch/stage_4/ge_block_1_repeat_1/gather_expansion_block/stride_equal_two_module/1x1_conv_block/conv/Conv2D LaneNet_1/bisenetv2_frontend/semantic_branch/stage_4/ge_block_1_repeat_1/gather_expansion_block/stride_equal_two_module/fused_features LaneNet_1/bisenetv2_frontend/semantic_branch/stage_4/ge_block_1_repeat_1/gather_expansion_block/stride_equal_two_module/ge_output LaneNet_1/bisenetv2_frontend/semantic_branch/stage_4/ge_block_2_repeat_1/gather_expansion_block/stride_equal_one_module/3x3_conv_block/conv/Conv2D LaneNet_1/bisenetv2_frontend/semantic_branch/stage_4/ge_block_2_repeat_1/gather_expansion_block/stride_equal_one_module/depthwise_conv_block/depthwise_conv_output LaneNet_1/bisenetv2_frontend/semantic_branch/stage_4/ge_block_2_repeat_1/gather_expansion_block/stride_equal_one_module/1x1_conv_block/conv/Conv2D LaneNet_1/bisenetv2_frontend/semantic_branch/stage_4/ge_block_1_repeat_1/gather_expansion_block/stride_equal_two_module/ge_output___tr4LaneNet_1/bisenetv2_frontend/semantic_branch/stage_4/ge_block_2_repeat_1/gather_expansion_block/stride_equal_one_module/fused_features LaneNet_1/bisenetv2_frontend/semantic_branch/stage_4/ge_block_2_repeat_1/gather_expansion_block/stride_equal_one_module/1x1_conv_block/bn/FusedBatchNorm___tr4LaneNet_1/bisenetv2_frontend/semantic_branch/stage_4/ge_block_2_repeat_1/gather_expansion_block/stride_equal_one_module/fused_features LaneNet_1/bisenetv2_frontend/semantic_branch/stage_4/ge_block_2_repeat_1/gather_expansion_block/stride_equal_one_module/fused_features LaneNet_1/bisenetv2_frontend/semantic_branch/stage_4/ge_block_2_repeat_1/gather_expansion_block/stride_equal_one_module/ge_output LaneNet_1/bisenetv2_frontend/semantic_branch/stage_4/ge_block_2_repeat_1/gather_expansion_block/stride_equal_one_module/ge_output___tr4LaneNet_1/bisenetv2_frontend/semantic_branch/stage_5/ge_block_1_repeat_1/gather_expansion_block/stride_equal_two_module/input_project_dw_conv_block/depthwise_conv_output LaneNet_1/bisenetv2_frontend/semantic_branch/stage_5/ge_block_1_repeat_1/gather_expansion_block/stride_equal_two_module/input_project_dw_conv_block/depthwise_conv_output LaneNet_1/bisenetv2_frontend/semantic_branch/stage_5/ge_block_1_repeat_1/gather_expansion_block/stride_equal_two_module/input_project_1x1_conv_block/conv/Conv2D LaneNet_1/bisenetv2_frontend/semantic_branch/stage_4/ge_block_2_repeat_1/gather_expansion_block/stride_equal_one_module/ge_output___tr4LaneNet_1/bisenetv2_frontend/semantic_branch/stage_5/ge_block_1_repeat_1/gather_expansion_block/stride_equal_two_module/3x3_conv_block/conv/Conv2D LaneNet_1/bisenetv2_frontend/semantic_branch/stage_5/ge_block_1_repeat_1/gather_expansion_block/stride_equal_two_module/3x3_conv_block/conv/Conv2D LaneNet_1/bisenetv2_frontend/semantic_branch/stage_5/ge_block_1_repeat_1/gather_expansion_block/stride_equal_two_module/depthwise_conv_block_1/depthwise_conv_output LaneNet_1/bisenetv2_frontend/semantic_branch/stage_5/ge_block_1_repeat_1/gather_expansion_block/stride_equal_two_module/depthwise_conv_block_2/depthwise_conv_output LaneNet_1/bisenetv2_frontend/semantic_branch/stage_5/ge_block_1_repeat_1/gather_expansion_block/stride_equal_two_module/1x1_conv_block/conv/Conv2D LaneNet_1/bisenetv2_frontend/semantic_branch/stage_5/ge_block_1_repeat_1/gather_expansion_block/stride_equal_two_module/fused_features LaneNet_1/bisenetv2_frontend/semantic_branch/stage_5/ge_block_1_repeat_1/gather_expansion_block/stride_equal_two_module/ge_output LaneNet_1/bisenetv2_frontend/semantic_branch/stage_5/ge_block_2_repeat_1/gather_expansion_block/stride_equal_one_module/3x3_conv_block/conv/Conv2D LaneNet_1/bisenetv2_frontend/semantic_branch/stage_5/ge_block_2_repeat_1/gather_expansion_block/stride_equal_one_module/depthwise_conv_block/depthwise_conv_output LaneNet_1/bisenetv2_frontend/semantic_branch/stage_5/ge_block_2_repeat_1/gather_expansion_block/stride_equal_one_module/1x1_conv_block/conv/Conv2D LaneNet_1/bisenetv2_frontend/semantic_branch/stage_5/ge_block_1_repeat_1/gather_expansion_block/stride_equal_two_module/ge_output___tr4LaneNet_1/bisenetv2_frontend/semantic_branch/stage_5/ge_block_2_repeat_1/gather_expansion_block/stride_equal_one_module/fused_features LaneNet_1/bisenetv2_frontend/semantic_branch/stage_5/ge_block_2_repeat_1/gather_expansion_block/stride_equal_one_module/1x1_conv_block/bn/FusedBatchNorm___tr4LaneNet_1/bisenetv2_frontend/semantic_branch/stage_5/ge_block_2_repeat_1/gather_expansion_block/stride_equal_one_module/fused_features LaneNet_1/bisenetv2_frontend/semantic_branch/stage_5/ge_block_2_repeat_1/gather_expansion_block/stride_equal_one_module/fused_features LaneNet_1/bisenetv2_frontend/semantic_branch/stage_5/ge_block_2_repeat_1/gather_expansion_block/stride_equal_one_module/ge_output LaneNet_1/bisenetv2_frontend/semantic_branch/stage_5/ge_block_2_repeat_1/gather_expansion_block/stride_equal_one_module/ge_output___tr4LaneNet_1/bisenetv2_frontend/semantic_branch/stage_5/ge_block_2_repeat_2/gather_expansion_block/stride_equal_one_module/3x3_conv_block/conv/Conv2D LaneNet_1/bisenetv2_frontend/semantic_branch/stage_5/ge_block_2_repeat_2/gather_expansion_block/stride_equal_one_module/3x3_conv_block/conv/Conv2D LaneNet_1/bisenetv2_frontend/semantic_branch/stage_5/ge_block_2_repeat_2/gather_expansion_block/stride_equal_one_module/depthwise_conv_block/depthwise_conv_output LaneNet_1/bisenetv2_frontend/semantic_branch/stage_5/ge_block_2_repeat_2/gather_expansion_block/stride_equal_one_module/1x1_conv_block/conv/Conv2D LaneNet_1/bisenetv2_frontend/semantic_branch/stage_5/ge_block_2_repeat_2/gather_expansion_block/stride_equal_one_module/1x1_conv_block/bn/FusedBatchNorm___tr4LaneNet_1/bisenetv2_frontend/semantic_branch/stage_5/ge_block_2_repeat_2/gather_expansion_block/stride_equal_one_module/fused_features LaneNet_1/bisenetv2_frontend/semantic_branch/stage_5/ge_block_2_repeat_2/gather_expansion_block/stride_equal_one_module/fused_features LaneNet_1/bisenetv2_frontend/semantic_branch/stage_5/ge_block_2_repeat_2/gather_expansion_block/stride_equal_one_module/ge_output LaneNet_1/bisenetv2_frontend/semantic_branch/stage_5/ge_block_2_repeat_2/gather_expansion_block/stride_equal_one_module/ge_output___tr4LaneNet_1/bisenetv2_frontend/semantic_branch/stage_5/ge_block_2_repeat_3/gather_expansion_block/stride_equal_one_module/3x3_conv_block/conv/Conv2D LaneNet_1/bisenetv2_frontend/semantic_branch/stage_5/ge_block_2_repeat_3/gather_expansion_block/stride_equal_one_module/3x3_conv_block/conv/Conv2D LaneNet_1/bisenetv2_frontend/semantic_branch/stage_5/ge_block_2_repeat_3/gather_expansion_block/stride_equal_one_module/depthwise_conv_block/depthwise_conv_output LaneNet_1/bisenetv2_frontend/semantic_branch/stage_5/ge_block_2_repeat_3/gather_expansion_block/stride_equal_one_module/1x1_conv_block/conv/Conv2D LaneNet_1/bisenetv2_frontend/semantic_branch/stage_5/ge_block_2_repeat_3/gather_expansion_block/stride_equal_one_module/1x1_conv_block/bn/FusedBatchNorm___tr4LaneNet_1/bisenetv2_frontend/semantic_branch/stage_5/ge_block_2_repeat_3/gather_expansion_block/stride_equal_one_module/fused_features LaneNet_1/bisenetv2_frontend/semantic_branch/stage_5/ge_block_2_repeat_3/gather_expansion_block/stride_equal_one_module/fused_features LaneNet_1/bisenetv2_frontend/semantic_branch/stage_5/ge_block_2_repeat_3/gather_expansion_block/stride_equal_one_module/ge_output LaneNet_1/bisenetv2_frontend/semantic_branch/stage_5/ce_block_3_repeat_1/context_embedding_block/global_avg_pooling LaneNet_1/bisenetv2_frontend/semantic_branch/stage_5/ce_block_3_repeat_1/context_embedding_block/global_avg_pooling___tr4LaneNet_1/bisenetv2_frontend/semantic_branch/stage_5/ce_block_3_repeat_1/context_embedding_block/bn/FusedBatchNorm LaneNet_1/bisenetv2_frontend/semantic_branch/stage_5/ce_block_3_repeat_1/context_embedding_block/bn/FusedBatchNorm LaneNet_1/bisenetv2_frontend/semantic_branch/stage_5/ce_block_3_repeat_1/context_embedding_block/conv_block_1/conv/Conv2D LaneNet_1/bisenetv2_frontend/semantic_branch/stage_5/ce_block_3_repeat_1/context_embedding_block/conv_block_1/relu___tr4LaneNet_1/bisenetv2_frontend/semantic_branch/stage_5/ce_block_3_repeat_1/context_embedding_block/fused_features LaneNet_1/bisenetv2_frontend/semantic_branch/stage_5/ce_block_3_repeat_1/context_embedding_block/fused_features LaneNet_1/bisenetv2_frontend/semantic_branch/stage_5/ce_block_3_repeat_1/context_embedding_block/fused_features___tr4LaneNet_1/bisenetv2_frontend/semantic_branch/stage_5/ce_block_3_repeat_1/context_embedding_block/final_conv_block/Conv2D LaneNet_1/bisenetv2_frontend/semantic_branch/stage_5/ce_block_3_repeat_1/context_embedding_block/final_conv_block/Conv2D LaneNet_1/bisenetv2_frontend/aggregation_branch/guided_aggregation_block/semantic_branch/3x3_conv_block/conv/Conv2D LaneNet_1/bisenetv2_frontend/aggregation_branch/guided_aggregation_block/semantic_branch/semantic_upsample_features LaneNet_1/bisenetv2_frontend/aggregation_branch/guided_aggregation_block/semantic_branch/semantic_upsample_sigmoid LaneNet_1/bisenetv2_frontend/aggregation_branch/guided_aggregation_block/detail_branch/1x1_conv_block/Conv2D___tr4LaneNet_1/bisenetv2_frontend/aggregation_branch/guided_aggregation_block/aggregation_features/guided_detail_features LaneNet_1/bisenetv2_frontend/aggregation_branch/guided_aggregation_block/semantic_branch/semantic_upsample_sigmoid___tr4LaneNet_1/bisenetv2_frontend/aggregation_branch/guided_aggregation_block/aggregation_features/guided_detail_features LaneNet_1/bisenetv2_frontend/aggregation_branch/guided_aggregation_block/aggregation_features/guided_detail_features LaneNet_1/bisenetv2_frontend/aggregation_branch/guided_aggregation_block/detail_branch/3x3_conv_block/conv/Conv2D LaneNet_1/bisenetv2_frontend/aggregation_branch/guided_aggregation_block/detail_branch/avg_pooling_block LaneNet_1/bisenetv2_frontend/aggregation_branch/guided_aggregation_block/semantic_branch/3x3_dw_conv_block/depthwise_conv_output LaneNet_1/bisenetv2_frontend/aggregation_branch/guided_aggregation_block/semantic_branch/1x1_conv_block/Conv2D LaneNet_1/bisenetv2_frontend/aggregation_branch/guided_aggregation_block/semantic_branch/semantic_remain_sigmoid LaneNet_1/bisenetv2_frontend/aggregation_branch/guided_aggregation_block/detail_branch/avg_pooling_block___tr4LaneNet_1/bisenetv2_frontend/aggregation_branch/guided_aggregation_block/aggregation_features/guided_semantic_features LaneNet_1/bisenetv2_frontend/aggregation_branch/guided_aggregation_block/semantic_branch/semantic_remain_sigmoid___tr4LaneNet_1/bisenetv2_frontend/aggregation_branch/guided_aggregation_block/aggregation_features/guided_semantic_features LaneNet_1/bisenetv2_frontend/aggregation_branch/guided_aggregation_block/aggregation_features/guided_semantic_features LaneNet_1/bisenetv2_frontend/aggregation_branch/guided_aggregation_block/aggregation_features/guided_semantic_features___tr4LaneNet_1/bisenetv2_frontend/aggregation_branch/guided_aggregation_block/aggregation_features/guided_upsample_features LaneNet_1/bisenetv2_frontend/aggregation_branch/guided_aggregation_block/aggregation_features/guided_upsample_features LaneNet_1/bisenetv2_frontend/aggregation_branch/guided_aggregation_block/aggregation_features/guided_upsample_features___tr4LaneNet_1/bisenetv2_frontend/aggregation_branch/guided_aggregation_block/aggregation_features/fused_features LaneNet_1/bisenetv2_frontend/aggregation_branch/guided_aggregation_block/aggregation_features/fused_features LaneNet_1/bisenetv2_frontend/aggregation_branch/guided_aggregation_block/aggregation_features/fused_features___tr4LaneNet_1/bisenetv2_frontend/aggregation_branch/guided_aggregation_block/aggregation_features/aggregation_feature_output/conv/Conv2D LaneNet_1/bisenetv2_frontend/aggregation_branch/guided_aggregation_block/aggregation_features/aggregation_feature_output/conv/Conv2D LaneNet_1/bisenetv2_frontend/binary_segmentation_branch/conv_3x3/conv/Conv2D LaneNet_1/bisenetv2_frontend/binary_segmentation_branch/conv_1x1/conv/Conv2D LaneNet_1/bisenetv2_frontend/binary_segmentation_branch/final_conv/conv/Conv2D LaneNet_1/bisenetv2_frontend/binary_segmentation_branch/binary_logits LaneNet_1/bisenetv2_frontend/binary_segmentation_branch/binary_logits___tr4LaneNet_1/bisenetv2_backend/binary_seg/Softmax LaneNet_1/bisenetv2_backend/binary_seg/Softmax LaneNet_1/bisenetv2_backend/binary_seg/ArgMax lanenet/Cast lanenet/final_binary_output LaneNet_1/bisenetv2_frontend/instance_segmentation_branch/conv_3x3/conv/Conv2D LaneNet_1/bisenetv2_frontend/instance_segmentation_branch/conv_1x1/conv/Conv2D LaneNet_1/bisenetv2_frontend/instance_segmentation_branch/instance_logits LaneNet_1/bisenetv2_backend/instance_seg/pix_bn/FusedBatchNorm LaneNet_1/bisenetv2_backend/instance_seg/pix_relu LaneNet_1/bisenetv2_backend/instance_seg/pix_embedding_conv/Conv2D LaneNet_1/bisenetv2_backend/instance_seg/pix_embedding_conv/Conv2D___tr4lanenet/final_pixel_embedding_output lanenet/final_pixel_embedding_output
但是,在安卓上运行runSessionWithCallBack()时,只能打印前26个tensor name,说明后面都没有运行,请问这是什么情况?
LaneNet_1/bisenetv2_frontend/detail_branch/stage_1/conv_block_1_repeat_1/3x3_conv/conv/Conv2D LaneNet_1/bisenetv2_frontend/detail_branch/stage_1/conv_block_2_repeat_1/3x3_conv/conv/Conv2D LaneNet_1/bisenetv2_frontend/detail_branch/stage_2/conv_block_1_repeat_1/3x3_conv/conv/Conv2D LaneNet_1/bisenetv2_frontend/detail_branch/stage_2/conv_block_2_repeat_1/3x3_conv/conv/Conv2D LaneNet_1/bisenetv2_frontend/detail_branch/stage_2/conv_block_2_repeat_2/3x3_conv/conv/Conv2D LaneNet_1/bisenetv2_frontend/detail_branch/stage_3/conv_block_1_repeat_1/3x3_conv/conv/Conv2D LaneNet_1/bisenetv2_frontend/detail_branch/stage_3/conv_block_2_repeat_1/3x3_conv/conv/Conv2D LaneNet_1/bisenetv2_frontend/detail_branch/stage_3/conv_block_2_repeat_2/3x3_conv/conv/Conv2D LaneNet_1/bisenetv2_frontend/aggregation_branch/guided_aggregation_block/detail_branch/3x3_dw_conv_block/depthwise_conv_output LaneNet_1/bisenetv2_frontend/aggregation_branch/guided_aggregation_block/detail_branch/1x1_conv_block/Conv2D LaneNet_1/bisenetv2_frontend/semantic_branch/stage_1/se_block_1_repeat_1/stem_block/conv_block_1/conv/Conv2D LaneNet_1/bisenetv2_frontend/semantic_branch/stage_1/se_block_1_repeat_1/stem_block/downsample_branch_left/1x1_conv_block/conv/Conv2D LaneNet_1/bisenetv2_frontend/semantic_branch/stage_1/se_block_1_repeat_1/stem_block/downsample_branch_left/3x3_conv_block/conv/Conv2D LaneNet_1/bisenetv2_frontend/semantic_branch/stage_1/se_block_1_repeat_1/stem_block/downsample_branch_right/maxpooling_block LaneNet_1/bisenetv2_frontend/semantic_branch/stage_1/se_block_1_repeat_1/stem_block/concate_features LaneNet_1/bisenetv2_frontend/semantic_branch/stage_1/se_block_1_repeat_1/stem_block/final_conv_block/conv/Conv2D LaneNet_1/bisenetv2_frontend/semantic_branch/stage_3/ge_block_1_repeat_1/gather_expansion_block/stride_equal_two_module/input_project_dw_conv_block/depthwise_conv_output LaneNet_1/bisenetv2_frontend/semantic_branch/stage_3/ge_block_1_repeat_1/gather_expansion_block/stride_equal_two_module/input_project_1x1_conv_block/conv/Conv2D LaneNet_1/bisenetv2_frontend/semantic_branch/stage_3/ge_block_1_repeat_1/gather_expansion_block/stride_equal_two_module/3x3_conv_block/conv/Conv2D LaneNet_1/bisenetv2_frontend/semantic_branch/stage_3/ge_block_1_repeat_1/gather_expansion_block/stride_equal_two_module/depthwise_conv_block_1/depthwise_conv_output LaneNet_1/bisenetv2_frontend/semantic_branch/stage_3/ge_block_1_repeat_1/gather_expansion_block/stride_equal_two_module/depthwise_conv_block_2/depthwise_conv_output LaneNet_1/bisenetv2_frontend/semantic_branch/stage_3/ge_block_1_repeat_1/gather_expansion_block/stride_equal_two_module/1x1_conv_block/conv/Conv2D LaneNet_1/bisenetv2_frontend/semantic_branch/stage_3/ge_block_1_repeat_1/gather_expansion_block/stride_equal_two_module/fused_features LaneNet_1/bisenetv2_frontend/semantic_branch/stage_3/ge_block_1_repeat_1/gather_expansion_block/stride_equal_two_module/ge_output LaneNet_1/bisenetv2_frontend/semantic_branch/stage_3/ge_block_2_repeat_1/gather_expansion_block/stride_equal_one_module/3x3_conv_block/conv/Conv2D LaneNet_1/bisenetv2_frontend/semantic_branch/stage_3/ge_block_2_repeat_1/gather_expansion_block/stride_equal_one_module/depthwise_conv_block/depthwise_conv_output
非常感谢您的开源项目,谢谢。从中学到很多,在看代码的过程中有点疑问,就是为什么在visualize_instance_segmentation_result中color_map有7类,而最后结果只显示0、1、2三类?
@MaybeShewill-CV Hi.
I am planning to use MNN-LaneNet with VULKAN backend. Is it possible to make this repo with VULKAN backend ? I read cmakelists.txt but there is no any thing about vulkan backend.
I built MNN project and used libMNN.so in my MNN_LaneNet project but MNN_LaneNet is slower when i use VULKAN backend.
with mnn_config.type = MNN_FORWARD_CPU result :
detect, 211, cost time: 0.246000 ms
detect, 146, cost time: 0.838000 ms
detect, 153, cost time: 872.728027 ms
detect, 204, cost time: 29.430002 ms
with mnn_config.type = MNN_FORWARD_VULKAN result :
Vulkan don't support 2, ArgMax: lanenet_model/vgg_backend/binary_seg/ArgMax
Vulkan don't support 9, Cast: lanenet/Cast
detect, 146, cost time: 0.755000 ms
detect, 153, cost time: 2487.019043 ms
detect, 204, cost time: 31.821001 ms
Hello,I don't know the MNNConverter_TOOL_PATH?I had make the mnn,but i don't find the tool_path?can you give me the examples?
python ./tools/freeze_lanenet_model.py it can output tusimple_lanenet.pb, I can parse this model file, and visualize well. but convert mnn model err:
==== output_shape is not consistent with inferred output shape in MNN. ====
====(height,width): (32,64) vs (1024,1024)
====Converte Tensorflow's Op lanenet_model/vgg_frontend/vgg16_decode_module/binary_seg_decode/decode_stage_4_fuse/deconv/deconv/conv2d_transpose , type = Conv2DBackpropInput, failed, may be some node is not const
==== output_shape is not consistent with inferred output shape in MNN. ====
====(height,width): (32,64) vs (1024,1024)
====Converte Tensorflow's Op lanenet_model/vgg_frontend/vgg16_decode_module/binary_seg_decode/decode_stage_4_fuse/deconv/deconv/conv2d_transpose , type = Conv2DBackpropInput, failed, may be some node is not const
==== output_shape is not consistent with inferred output shape in MNN. ====
====(height,width): (32,64) vs (1024,1024)
====Converte Tensorflow's Op lanenet_model/vgg_frontend/vgg16_decode_module/binary_seg_decode/decode_stage_4_fuse/deconv/deconv/conv2d_transpose , type = Conv2DBackpropInput, failed, may be some node is not const
1=1 in onComputeInfo, 389
How should i do, thx !
A declarative, efficient, and flexible JavaScript library for building user interfaces.
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
An Open Source Machine Learning Framework for Everyone
The Web framework for perfectionists with deadlines.
A PHP framework for web artisans
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
Some thing interesting about web. New door for the world.
A server is a program made to process requests and deliver data to clients.
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
Some thing interesting about visualization, use data art
Some thing interesting about game, make everyone happy.
We are working to build community through open source technology. NB: members must have two-factor auth.
Open source projects and samples from Microsoft.
Google ❤️ Open Source for everyone.
Alibaba Open Source for everyone
Data-Driven Documents codes.
China tencent open source team.