Before, Frame Processors ran on a separate Thread.
After, Frame Processors run fully synchronous and always at the same FPS as the Camera.
Two new functions have been introduced:
* `runAtTargetFps(fps: number, func: () => void)`: Runs the given code as often as the given `fps`, effectively throttling it's calls.
* `runAsync(frame: Frame, func: () => void)`: Runs the given function on a separate Thread for Frame Processing. A strong reference to the Frame is held as long as the function takes to execute.
You can use `runAtTargetFps` to throttle calls to a specific API (e.g. if your Camera is running at 60 FPS, but you only want to run face detection at ~25 FPS, use `runAtTargetFps(25, ...)`.)
You can use `runAsync` to run a heavy algorithm asynchronous, so that the Camera is not blocked while your algorithm runs. This is useful if your main sync processor draws something, and your async processor is doing some image analysis on the side.
You can also combine both functions.
Examples:
```js
const frameProcessor = useFrameProcessor((frame) => {
'worklet'
console.log("I'm running at 60 FPS!")
}, [])
```
```js
const frameProcessor = useFrameProcessor((frame) => {
'worklet'
console.log("I'm running at 60 FPS!")
runAtTargetFps(10, () => {
'worklet'
console.log("I'm running at 10 FPS!")
})
}, [])
```
```js
const frameProcessor = useFrameProcessor((frame) => {
'worklet'
console.log("I'm running at 60 FPS!")
runAsync(frame, () => {
'worklet'
console.log("I'm running on another Thread, I can block for longer!")
})
}, [])
```
```js
const frameProcessor = useFrameProcessor((frame) => {
'worklet'
console.log("I'm running at 60 FPS!")
runAtTargetFps(10, () => {
'worklet'
runAsync(frame, () => {
'worklet'
console.log("I'm running on another Thread at 10 FPS, I can block for longer!")
})
})
}, [])
```
* feat: Allow returning of ImageProxy in a Frame Processor
* chore: Clean up
* fix: Missing space
* Update useFrameProcessor.ts
* Revert "Update useFrameProcessor.ts"
This reverts commit 9c645489cdfdf2079972669756a2cd20cc81e25e.
* Calculate a format's video dimensions based on supported resolutions and photo dimensions
* Add Android fallback strategy for recording quality
* Ensure that session props are not ignored when app is resumed
* Simplify setting Android video dimensions in format
* Modify Android imageAnalysisBuilder to use photoSize
* Update onHostResume function to reference android preview issue
* Add missing Android capture errors
Accessing previewView.bitmap was throwing an error because it wasn't being done on the main thread.
Any access to previewView needs to be done on the main (UI) thread. This commit fixes the issue by
ensuring this access is now run on the main thread.
Fixes#547
This commit fixes#758. I was having the same issue and looked into it a bit. I found
[this StackOverflow answer](https://stackoverflow.com/a/60585382) which described a
solution to the same problem. Rather than manually calculate the focus point, we can
get the PreviewView to do it for us. This fixes the issue because the PreviewView
factors in any scaling or resizing of the view on the screen, which we weren't doing
before. The only potential issue is that this needs to run on the UI thread
(which is what the `withContext` is doing), but I've tested it with frame processors
enabled and disabled, and have found no issues in either case.
* fix: Use `rootDir` instead of `projectDir`
* Revert "fix: Use `rootDir` instead of `projectDir`"
This reverts commit 058e0a110bcf9b688e12a1cccbac2f23a29fa55c.
* fix: Find node_modules path where react-native/ lives
* fix: Figure out VisionCameraExample project
* Revert "fix: Figure out VisionCameraExample project"
This reverts commit 7ca455098244dd62280d40586062803d1ccc2c5f.
Technically `JHashMap` is duplicated now, but in separate namespaces. If I were to remove my `JHashMap` (and `JArrayList`) definitions, the user is forced to use fbjni v3.
* Add custom `onViewReady` event to get layout
`componentDidMount` is async, so the native view _might_ not exist yet causing a race condition in the `setFrameProcessor` code.
This PR fixes this by calling `setFrameProcessor` only after the native view has actually mounted, and to ensure that I created a custom event that fires at that point.
* Update CameraView.swift
* chore: Upgrade CameraX to alpha09
* Remove custom ProGuard file
It's no longer needed, CameraX now ships one.
* Set `targetSdkVersion` to `31`
* set `compileSdkVersion` to 31
* Add `android:exported=false`
* De-allocate `frame` HybridClass with JNI class loader if using Hermes
See 1b3a0c2612
* Don't wrap in `#if FOR_HERMES`, other `jsi::Runtime`s might also run GC on another Thread.
* Use `jni::local_ref` for `FrameHostObject`
* Update JImageProxyHostObject.cpp
* Only run with JNI `ClassLoader` if ctor Thread ID != dtor Thread ID
* Upgrade reanimated to 2.3.0-beta.1 to fix JNI crash
* Remove `this_thread::get_id()`
* Update Podfile.lock
