{"id":50759973,"url":"https://github.com/amd/imagetilingnpu","last_synced_at":"2026-06-11T09:01:10.192Z","repository":{"id":360472678,"uuid":"1068502037","full_name":"amd/ImageTilingNPU","owner":"amd","description":null,"archived":false,"fork":false,"pushed_at":"2026-05-26T14:02:53.000Z","size":17,"stargazers_count":5,"open_issues_count":2,"forks_count":2,"subscribers_count":0,"default_branch":"main","last_synced_at":"2026-05-26T16:07:58.998Z","etag":null,"topics":[],"latest_commit_sha":null,"homepage":null,"language":"C++","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":"other","status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/amd.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":null,"funding":null,"license":"LICENSE.txt","code_of_conduct":null,"threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":null,"support":null,"governance":null,"roadmap":null,"authors":null,"dei":null,"publiccode":null,"codemeta":null,"zenodo":null,"notice":null,"maintainers":null,"copyright":null,"agents":null,"dco":null,"cla":null}},"created_at":"2025-10-02T13:35:05.000Z","updated_at":"2026-05-26T14:24:56.000Z","dependencies_parsed_at":null,"dependency_job_id":null,"html_url":"https://github.com/amd/ImageTilingNPU","commit_stats":null,"previous_names":["amd/imagetilingnpu"],"tags_count":null,"template":false,"template_full_name":null,"purl":"pkg:github/amd/ImageTilingNPU","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/amd%2FImageTilingNPU","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/amd%2FImageTilingNPU/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/amd%2FImageTilingNPU/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/amd%2FImageTilingNPU/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/amd","download_url":"https://codeload.github.com/amd/ImageTilingNPU/tar.gz/refs/heads/main","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/amd%2FImageTilingNPU/sbom","scorecard":null,"host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":286080680,"owners_count":34190585,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2026-05-26T15:22:16.424Z","status":"online","status_checked_at":"2026-06-11T02:00:06.485Z","response_time":57,"last_error":null,"robots_txt_status":"success","robots_txt_updated_at":"2025-07-24T06:49:26.215Z","robots_txt_url":"https://github.com/robots.txt","online":true,"can_crawl_api":true,"host_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub","repositories_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories","repository_names_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repository_names","owners_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners"}},"keywords":[],"created_at":"2026-06-11T09:01:09.020Z","updated_at":"2026-06-11T09:01:10.193Z","avatar_url":"https://github.com/amd.png","language":"C++","funding_links":[],"categories":[],"sub_categories":[],"readme":"# ImageTilingNPU\n\n## Project introduction\n\nThis repository provides an implementation for deploying neural network-based image super-resolution models on the AMD XDNA\u003csup\u003eTM\u003c/sup\u003e platform, leveraging the NPU(Neural Processing Unit) for efficient inference. The goal is to enable fast, low-power, and high-quality super-resolution on edge and embedded device powered by AMD XDNA\u003csup\u003eTM\u003c/sup\u003e.\n\n## Features\n\n- High-Quality Super-Resolution: Supports neural network models for enhancing low-resolution images to high-resolution.\n- AMD XDNA NPU Acceleration: Optimized for inference on AMD XDNA\u003csup\u003eTM\u003c/sup\u003e Neural Processing Unit for low-latency and energy-efficient execution.\n- Model Support: Both ONNX and Xmodel are supported\n- Lightweight \u0026 Efficient: Minimal dependencies and optimized code for fast inference on edge devices.\n\n## Requirements\n\n- Windows platform with AMD XDNA\u003csup\u003eTM\u003c/sup\u003e NPU hardware and NPU driver installed\n- Visual Studio 2022\n- Cmake \u003e 3.4\n\n## Build\n\n1. Prepare onnxruntime, download onnxruntime headers and libraries from [onnxruntime OMS](https://download.amd.com/opendownload/RyzenAI/NPU_image_tiling/imagetiling_dep.zip)\n\n2. In CMakeLists.txt, set the ORT_DIR variable to the onnxruntime path in onnxruntime OMS.\n\n3. Copy the third_party folder in onnxruntime OMS to project root path.\n\n4. Download opencv from https://github.com/opencv/opencv/releases/download/4.11.0/opencv-4.11.0-windows.exe, then copy the opencv2 headers and opencv_world4110.lib library to third_party folder.\n\n5. Build the project\n   \n   ```\n   cmake -B build -S .\n   cmake --build build --config release\n   ```\n\n6. Copy build/*.exe opencv/build/x64/vc16/bin/opencv_world4110.dll onnxruntime/bin/*.dll to bin folder\n\n## Example\n\nCopy AMD_AIE2P_8x4x1_Overlay_CFG5.xclbin to ${XCLBIN_PATH}(any dir is okay)\n\n### Functional test\n\nWe provided two 2 ways to implement super-resolution, both onnx and xmodel\n\nFor onnx model\n\n```\ntest_SR_onnx.exe --model input.onnx --vai_options \"xclbin|${XCLBIN_PATH}/AMD_AIE2P_8x4x1_Overlay_CFG5.xclbin config_file|config_file/vaip_config.json\" --image input.png\n```\n\nFor xmodel model\n\n```\ntest_SR_xmodel.exe --model input.xmodel --vai_options \"xclbin|${XCLBIN_PATH}/AMD_AIE2P_8x4x1_Overlay_CFG5.xclbin\" --image input.png\n```\n\n### Performance test\n\nNavigate to NPU driver path\n\n```\nxrt-smi configure --pmode performance\n```\n\nFor onnx model\n\n```\nperf_SR_onnx.exe --model input.onnx --image_height 1440 --image_width 2560 --vai_options \"xclbin|${XCLBIN_PATH}/AMD_AIE2P_8x4x1_Overlay_CFG5.xclbin ctx_idx|1\"\n```\n\nFor xmodel model\n\n```\nperf_SR_xmodel.exe --model input.xmodel --image_height 1440 --image_width 2560 --vai_options \"xclbin|${XCLBIN_PATH}/AMD_AIE2P_8x4x1_Overlay_CFG5.xclbin ctx_idx|1\"\n```\n\n## Contribute code\n\nIf you meet any problem or request a new feature, you're welcome to create an issue.\nIf you can solve any of the issues, you're welcome to send the PR to us.","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Famd%2Fimagetilingnpu","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Famd%2Fimagetilingnpu","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Famd%2Fimagetilingnpu/lists"}