Any chance of using imagebuilder on aarch64?

I am trying to set up Imagebuilder to build multiple images with customized uci-defaults for every one of them. I have functional build on x86_64 but when I transfer it to Raspi with aarch64 architecture, it doesn't work with following results.

/opt/cudyflasher/imagebuilder/staging_dir/host/bin/find: line 5: /opt/cudyflasher/imagebuilder/staging_dir/host/bin/../lib/ld-linux-x86-64.so.2: cannot execute binary file: Exec format error
/opt/cudyflasher/imagebuilder/staging_dir/host/bin/find: line 5: /opt/cudyflasher/imagebuilder/staging_dir/host/bin/../lib/ld-linux-x86-64.so.2: Success
/opt/cudyflasher/imagebuilder/staging_dir/host/bin/opkg: line 5: /opt/cudyflasher/imagebuilder/staging_dir/host/bin/../lib/ld-linux-x86-64.so.2: cannot execute binary file: Exec format error
/opt/cudyflasher/imagebuilder/staging_dir/host/bin/opkg: line 5: /opt/cudyflasher/imagebuilder/staging_dir/host/bin/../lib/ld-linux-x86-64.so.2: Success

Is there any way to tell make to cross-compile these binaries to aarch64 architecture during compilation process or any other approach to get functional imagebuilder on another architecture?

One option is to build imagebuilder natively
Other option: https://wiki.debian.org/QemuUserEmulation

The imagebuilder, as provided by the OpenWrt project, is currently x86_64-only. If you build it from source on an arm64 host yourself, you could build it as arm64 binaries (there may be some bugs hiding, as this isn't tested, but in general it should work), however in that case it might be more sensible to build whatever you want to build from source to begin with (unless you have a specific need to use the imagebuilder).

While approaches via qemu or qemu-arm64-static might be successful, they're hardly sensible for this use case, considering that even booting an x86_64 live linux from a USB stick (virtualized or natively) would be easier and faster.