I've experimented quite a bit with the new API designed to neutralize image colors using the iPhone flash, and I think the concept is brilliant. The flash could potentially serve as a substitute for a color checker, given our full control over it. However, I believe there are several areas where this API could be improved.
Firstly, the resulting images often appear "unattractive"—colors tend to look faded, and the images themselves can be overly bright and washed out, losing the natural ambiance, shadows, and introducing unwanted flash reflections. There is also inconsistency in color rendering; for example, yellows sometimes appear unnatural, possibly due to reflections. In some cases, all the colors in the image are completely desaturated or become black and white if another light source does not fully illuminate the scene. Additionally, the shadows cast by the flash don't correct the colors properly since they fall outside the flash's range.
I think many of these issues could be resolved if we had access to ProRAW images capturing both the ambient light (without flash) and the flash-illuminated scene. With these, we could use specific colors in the image as a reference, similar to a color checker, to create an ICC profile or color transformation matrix to adjust the image colors more globally. This approach could help retain the shadows from the ambient light while still correcting colors to a neutral tone.
Access to ProRAW data is crucial for this, as it would provide images without the saturation issues that can affect some colors and with a linear tone curve. I hope this suggestion makes sense and could help improve the API's effectiveness.
Post
Replies
Boosts
Views
Activity
Hi everyone, I attempted to use the MultivariateLinearRegressor from the Create ML Components framework to fit some multi-dimensional data linearly (4 dimensions in my example). I aim to obtain multi-dimensional output points (2 points in my example). However, when I fit the model with my training data and test it, it appears that only the first element of my training data is used for training, regardless of whether I use CreateMLComponents.AnnotatedBatch or [CreateMLComponents.AnnotatedFeature, CoreML.MLShapedArray>] as input.
let sourceMatrix: [[Double]] = [
[0,0.1,0.2,0.3],
[0.5,0.2,0.6,0.2]
]
let referenceMatrix: [[Double]] = [
[0.2,0.7],
[0.9,0.1]
]
Here is a test code to test the function (ios 18.0 beta, Xcode 16.0 beta)
In this example I train the model to learn 2 multidimensional points (4 dimensions) and here are the results of the predictions:
▿ 2 elements
▿ 0 : AnnotatedPrediction<MLShapedArray<Double>, MLShapedArray<Double>>
▿ prediction : 0.20000000298023224 0.699999988079071
▿ _storage : <StandardStorage<Double>: 0x600002ad8270>
▿ annotation : 0.2 0.7
▿ _storage : <StandardStorage<Double>: 0x600002b30600>
▿ 1 : AnnotatedPrediction<MLShapedArray<Double>, MLShapedArray<Double>>
▿ prediction : 0.23158159852027893 0.9509953260421753
▿ _storage : <StandardStorage<Double>: 0x600002ad8c90>
▿ annotation : 0.9 0.1
▿ _storage : <StandardStorage<Double>: 0x600002b55f20>
0.23158159852027893 0.9509953260421753 is totally random and should be far more closer to [0.9,0.1].
Here is the test code : ( i run it on "My mac, Designed for Ipad")
ContentView.swift
import CoreImage
import CoreImage.CIFilterBuiltins
import UIKit
import CoreGraphics
import Accelerate
import Foundation
import CoreML
import CreateML
import CreateMLComponents
func createMLShapedArray(from array: [Double], shape: [Int]) -> MLShapedArray<Double> {
return MLShapedArray<Double>(scalars: array, shape: shape)
}
func calculateTransformationMatrixWithNonlinearity(sourceRGB: [[Double]], referenceRGB: [[Double]], degree: Int = 3) async throws -> MultivariateLinearRegressor<Double>.Model {
let annotatedFeatures2 = zip(sourceRGB, referenceRGB).map { (featureArray, targetArray) -> AnnotatedFeature<MLShapedArray<Double>, MLShapedArray<Double>> in
let featureMLShapedArray = createMLShapedArray(from: featureArray, shape: [featureArray.count])
let targetMLShapedArray = createMLShapedArray(from: targetArray, shape: [targetArray.count])
return AnnotatedFeature(feature: featureMLShapedArray, annotation: targetMLShapedArray)
}
// Flatten the sourceRGBPoly into a single-dimensional array
var flattenedArray = sourceRGB.flatMap { $0 }
let featuresMLShapedArray = createMLShapedArray(from: flattenedArray, shape: [2, 4])
flattenedArray = referenceRGB.flatMap { $0 }
let targetMLShapedArray = createMLShapedArray(from: flattenedArray, shape: [2, 2])
// Create AnnotatedFeature instances
/* let annotatedFeatures2: [AnnotatedFeature<MLShapedArray<Double>, MLShapedArray<Double>>] = [
AnnotatedFeature(feature: featuresMLShapedArray, annotation: targetMLShapedArray)
]*/
let annotatedBatch = AnnotatedBatch(features: featuresMLShapedArray, annotations: targetMLShapedArray)
var regressor = MultivariateLinearRegressor<Double>()
regressor.configuration.learningRate = 0.1
regressor.configuration.maximumIterationCount=5000
regressor.configuration.batchSize=2
let model = try await regressor.fitted(to: annotatedBatch,validateOn: nil)
//var model = try await regressor.fitted(to: annotatedFeatures2)
// Proceed to prediction once the model is fitted
let predictions = try await model.prediction(from: annotatedFeatures2)
// Process or use the predictions
print(predictions)
print("Predictions:", predictions)
return model
}
struct ContentView: View {
var body: some View {
VStack {}
.onAppear {
Task {
do {
let sourceMatrix: [[Double]] = [
[0,0.1,0.2,0.3],
[0.5,0.2,0.6,0.2]
]
let referenceMatrix: [[Double]] = [
[0.2,0.7],
[0.9,0.1]
]
let model = try await calculateTransformationMatrixWithNonlinearity(sourceRGB: sourceMatrix, referenceRGB: referenceMatrix, degree: 2
)
print("Model fitted successfully:", model)
} catch {
print("Error:", error)
}
}
}
}
}