fabio-sim/LightGlue-ONNX

Issues

• Could you tell me your versions of tensorrt and cuDNN? I encountered compatibility issues, thank you

  #96 opened 2 months ago
  2

• TracerWarning: torch.tensor results are registered as constants in the trace.

  #95 opened 2 months ago
  1

• OPENVINO

  #93 opened 2 months ago
  2

• match extractor_type

  #92 opened 2 months ago
  1

• Lightglue produces wrong matches after TensorRT optimization

  #90 opened 2 months ago
  2

• QKV Error when keypoint number is small

  #89 opened 2 months ago
  1

• onnx vs pth performance

  #88 opened 3 months ago
  3

• Match indices out of range

  #86 opened 4 months ago
  2

• Convert 2.0 End-to-end parallel ONNX file to TensorRT engine

  #85 opened 3 months ago
  4

• support for different sizes of inputs

  #83 opened 3 months ago
  3

• Key points from thesuperpoint

  #81 opened 3 months ago
  3

• the possiblity of supporting batch input

  #80 opened 5 months ago
  12

• error when using TensorrtExecutionProvider for ONNX models

  #78 opened 6 months ago
  18

• convert SuperPoint from onnx to engine

  #77 opened 5 months ago
  1

• False Positive Keypoints on uniform Images

  #76 opened 6 months ago
  3

• UnsupportedOperatorError: Exporting the operator 'aten::scaled_dot_product_attention' to ONNX opset version 17

  #75 opened 7 months ago
  3

• LightGlue

  #73 opened 5 months ago
  0

• how to convert onnx into an rknn model？

  #72 opened 7 months ago
  8

• Adding support for SIFT

  #71 opened 8 months ago
  3

• ".trt.onnx"export example

  #70 opened 6 months ago
  1

• ALIKED Support

  #69 opened 9 months ago
  2

• some problem that delopy on device, such as snapdragon snpe.

  #68 opened 6 months ago
  1

• Error when building the end2end onnx file.

  #67 opened 5 months ago
  1

• Internal Error (/lightglue/ArgMax)

  #66 opened 8 months ago
  4

• error when convert tensorRt engine model

  #65 opened 5 months ago
  4

• Converting a trained model to ONNX

  #63 opened 8 months ago
  4

• Result diffrent from the original repository?

  #62 opened 10 months ago
  3

• Running inference throws a CUDA exception

  #61 opened 5 months ago
  1

• Running inference using exported models in C++ very unstable/non-deterministic

  #60 opened 8 months ago
  2

• The output shape of lightglue's onnx model is dynamic. Does tensorrt support dynamic output?

  #59 opened a year ago
  4

• Build Superpoint onnx file to Tensorrt engine failed, encounter Error (Could not find any implementation for node {ForeignNode[/Flatten.../Transpose_3]}

  #58 opened a year ago
  5

• Is the lightglue run in the tensorrt mode, and how to using C++ inferface build engine?

  #56 opened a year ago
  6

• aliked

  #55 opened a year ago
  2

• support dynamic batch size

  #54 opened a year ago
  1

• jetson

  #53 opened 8 months ago
  1

• [softMaxV2Runner.cpp::execute::226] Error Code 1: Cask (shader run failed)

  #52 opened 8 months ago
  3

• NOT_IMPLEMENTED : Non-zero status code returned while running MultiHeadAttention node

  #51 opened a year ago
  3

• 我在jetson agx上运行infer.py，花费19s才完成，是我的环境有问题吗？tensorrt的版本一定要是8.6吗

  #50 opened 8 months ago
  1

• How do I export without specifying an img size

  #49 opened a year ago
  1

• Exporting the operator 'aten::scaled_dot_product_attention' to ONNX opset version 17 is not supported.

  #45 opened a year ago
  3

• Can't run on CPU

  #44 opened a year ago
  2

• Failed to test the end-to-end onnx model by trtexec

  #43 opened a year ago
  7

• Issues with the deployment on web

  #42 opened a year ago
  3

• 推理时间问题

  #41 opened a year ago
  5

• Integration to kornia?

  #40 opened a year ago
  6

• Support for open version of Superpoint

  #39 opened a year ago
  8

• too many values to unpack (expected 2)

  #38 opened a year ago
  1

• What's the difference between the relased models?

  #37 opened a year ago
  1

• use trtexec

  #36 opened a year ago
  4

• How to export FP32 model?

  #35 opened a year ago
  2