[Bug]: [NPU] CLRNet can't run on NPU with shape issue

[wdkwyf]

OpenVINO Version

2025.1

Operating System

Ubuntu 20.04 (LTS)

Device used for inference

NPU

Framework

ONNX

Model used

CLRNet

Issue description

benchmark_app -d NPU -t 1 -m /home/intel/tusimple_res18.onnx will cause results type error

[Step 7/11] Loading the model to the device
[ERROR] 10:16:26.252 [vpux-compiler] Got Diagnostic at loc(fused<{name = "/heads/Slice_16", type = "StridedSlice"}>["/heads/Slice_16"]) : Sequence lengths input size 1 is not equal to batch axis dimension of data input 22
loc(fused<{name = "/heads/Slice_16", type = "StridedSlice"}>["/heads/Slice_16"]): error: Sequence lengths input size 1 is not equal to batch axis dimension of data input 22
LLVM ERROR: Failed to infer result type(s).

NPU Driver Version: https://github.com/intel/linux-npu-driver/releases/tag/v1.16.0
Model Path: https://intel-my.sharepoint.com/:u:/p/yufei_wu/Eayj54X4nhhDoFBNRWY14bYB7Rvv9RCBg4deJMtkEaO7jg?e=wct2ZL

Step-by-step reproduction

benchmark_app -d NPU -t 1 -m /home/intel/tusimple_res18.onnx

Relevant log output

ov-llm-bench-env) yufei@ubuntu:~$ benchmark_app -d NPU -t 1 -m "/home/intel/tusimple_res18.onnx"
[Step 1/11] Parsing and validating input arguments
[ INFO ] Parsing input parameters
[Step 2/11] Loading OpenVINO Runtime
[ INFO ] OpenVINO:
[ INFO ] Build ................................. 2025.2.0-19070-2d384751f7c
[ INFO ]
[ INFO ] Device info:
[ INFO ] NPU
[ INFO ] Build ................................. 2025.2.0-19070-2d384751f7c
[ INFO ]
[ INFO ]
[Step 3/11] Setting device configuration
[ WARNING ] Performance hint was not explicitly specified in command line. Device(NPU) performance hint will be set to PerformanceMode.THROUGHPUT.
[Step 4/11] Reading model files
[ INFO ] Loading model files
[ INFO ] Read model took 181.05 ms
[ INFO ] Original model I/O parameters:
[ INFO ] Model inputs:
[ INFO ]     input (node: input) : f32 / [...] / [22,3,320,800]
[ INFO ] Model outputs:
[ INFO ]     2843 (node: 2843) : f32 / [...] / [22,192,78]
[Step 5/11] Resizing model to match image sizes and given batch
[ INFO ] Model batch size: 22
[Step 6/11] Configuring input of the model
[ INFO ] Model inputs:
[ INFO ]     input (node: input) : u8 / [N,C,H,W] / [22,3,320,800]
[ INFO ] Model outputs:
[ INFO ]     2843 (node: 2843) : f32 / [...] / [22,192,78]
[Step 7/11] Loading the model to the device
[ERROR] 10:16:26.252 [vpux-compiler] Got Diagnostic at loc(fused<{name = "/heads/Slice_16", type = "StridedSlice"}>["/heads/Slice_16"]) : Sequence lengths input size 1 is not equal to batch axis dimension of data input 22
loc(fused<{name = "/heads/Slice_16", type = "StridedSlice"}>["/heads/Slice_16"]): error: Sequence lengths input size 1 is not equal to batch axis dimension of data input 22
LLVM ERROR: Failed to infer result type(s).
Aborted (core dumped)

Issue submission checklist

• I'm reporting an issue. It's not a question.
• I checked the problem with the documentation, FAQ, open issues, Stack Overflow, etc., and have not found a solution.
• There is reproducer code and related data files such as images, videos, models, etc.