mmdet's resnet50 converted to onnx and then to haf reports an error

Hello, I have a model whose backbone is resnet50. I converted the backbone of the model to onnx, and the tests were all successful. When I converted to haf, I encountered the following problem.
Could you please help me find a solution? I have been stuck for a week. Thank you.

[warning] Cannot use graphviz, so no visualizations will be created
:white_check_mark: Initialized ClientRunner with hardware architecture: hailo8
:white_check_mark: ONNX model file found: /home/ubuntu/ycy/Sparse4D-3.0/onnx_models/backbone_neck2.onnx
[info] Translation started on ONNX model backbone_neck2
[info] Restored ONNX model backbone_neck2 (completion time: 00:00:00.27)
[info] Extracted ONNXRuntime meta-data for Hailo model (completion time: 00:00:01.71)
[info] Simplified ONNX model for a parsing retry attempt (completion time: 00:00:04.40)
:cross_mark: Failed to translate ONNX model to HAR: Invalid kernel shape for base conv layer base_conv2 (translated from /backbone/layer1/layer1.0/conv1/Conv).
Either the input shape doesn’t match the kernel shape, or the calculated groups number doesn’t match the expected ratio between kernel shape and input shape.
Kernel features: 384 Input features: 64 Groups: 0