Customize an image

You can customize images using properties on an Image composable (contentScale, colorFilter). You can also apply existing modifiers to apply different effects to your Image. Modifiers can be used on any composable, not just the Image composable, whereas contentScale and colorFilter are explicit parameters on the Image composable.

Content scale

Specify a contentScale option to crop or change how an image is scaled inside its bounds. By default, if you don't specify a contentScale option, ContentScale.Fit is used.

In the following example, the Image composable is restricted to a 150dp size with a border, and the background is set to yellow on the Image composable to showcase the different ContentScale options in the table below.

val imageModifier = Modifier
    .size(150.dp)
    .border(BorderStroke(1.dp, Color.Black))
    .background(Color.Yellow)
Image(
    painter = painterResource(id = R.drawable.dog),
    contentDescription = stringResource(id = R.string.dog_content_description),
    contentScale = ContentScale.Fit,
    modifier = imageModifier
)

Setting different ContentScale options results in different outputs. The following table helps you choose the correct ContentScale mode:

Source image The portrait source, which shows a dog. The landscape source, which shows a different dog.
ContentScale Result - Portrait Image: Result - Landscape Image:
ContentScale.Fit: Scale the image uniformly, keeping the aspect ratio (default). If content is smaller than the size, the image is scaled up to fit the bounds. A dog portrait scaled uniformly. A dog landscape scaled uniformly.
ContentScale.Crop: Center crop the image into the available space. A portrait image cropped to fill a square frame, showing only the central part of the image. A landscape image cropped to fill a square frame, showing only the central part of the image.
ContentScale.FillHeight: Scale the source maintaining the aspect ratio so that the bounds match the destination height. A portrait image scaled to fill the height of a square frame, showing the full image with yellow background visible on the left and right. A landscape image scaled to fill the height of a square frame, with the sides cropped.
ContentScale.FillWidth: Scale the source maintaining the aspect ratio so that the bounds match the destination width. A portrait image scaled to fill the width of a square frame, with the top and bottom cropped. A landscape image scaled to fill the width of a square frame, showing the full image with yellow background visible on the top and bottom.
ContentScale.FillBounds: Scale the content vertically and horizontally non-uniformly to fill the destination bounds. (Note: This distorts images if you place them in containers that don't match the exact ratio of the image). A portrait image distorted to completely fill a square frame, stretching the image. A landscape image distorted to completely fill a square frame, stretching the image.
ContentScale.Inside: Scale the source to maintain the aspect ratio inside the destination bounds. If the source is smaller than or equal to the destination in both dimensions, it behaves similarly to None. Content will always be contained within the bounds. If content is smaller than bounds, no scaling will apply. Source image bigger than bounds: A portrait image, originally larger than its square bounds, scaled down to fit while maintaining aspect ratio, showing yellow background on the sides. Source image smaller than bounds: A portrait image, originally smaller than its square bounds, displayed at its original size within the frame, showing a yellow background around it. Source image bigger than bounds: A landscape image, originally larger than its square bounds, scaled down to fit while maintaining aspect ratio, showing yellow background on the top and bottom. Source image smaller than bounds: A landscape image, originally smaller than its square bounds, displayed at its original size within the frame, showing a yellow background around it.
ContentScale.None: Don't apply any scaling to the source. If the content is smaller than destination bounds, it won't be scaled up to fit the area. Source image bigger than bounds: A portrait image, originally larger than its square bounds, displayed at its original size, with parts extending beyond the top and bottom of the frame. Source image smaller than bounds: A portrait image, originally smaller than its square bounds, displayed at its original size within the frame, showing a yellow background around it. Source image bigger than bounds: A landscape image, originally larger than its square bounds, displayed at its original size, with parts extending beyond the left and right of the frame. Source image smaller than bounds: A landscape image, originally smaller than its square bounds, displayed at its original size within the frame, showing a yellow background around it.

Clip an Image composable to a shape

To make an image fit into a shape, use the built-in clip modifier. To crop an image into a circle shape, use Modifier.clip(CircleShape):

Image(
    painter = painterResource(id = R.drawable.dog),
    contentDescription = stringResource(id = R.string.dog_content_description),
    contentScale = ContentScale.Crop,
    modifier = Modifier
        .size(200.dp)
        .clip(CircleShape)
)

An image of a dog clipped into a perfect circle.
Figure 1. Clipping an image with CircleShape.

For a rounded corner shape, use Modifier.clip(RoundedCornerShape(16.dp)), with the size of the corners you want to be rounded:

Image(
    painter = painterResource(id = R.drawable.dog),
    contentDescription = stringResource(id = R.string.dog_content_description),
    contentScale = ContentScale.Crop,
    modifier = Modifier
        .size(200.dp)
        .clip(RoundedCornerShape(16.dp))
)

A square image of a dog clipped with rounded corners.
Figure 2. Clipping an image with RoundedCornerShape.

You can also create your own clipping shape by extending Shape, and providing a Path for the shape to clip around:

class SquashedOval : Shape {
    override fun createOutline(
        size: Size,
        layoutDirection: LayoutDirection,
        density: Density
    ): Outline {
        val path = Path().apply {
            // We create an Oval that starts at ¼ of the width, and ends at ¾ of the width of the container.
            addOval(
                Rect(
                    left = size.width / 4f,
                    top = 0f,
                    right = size.width * 3 / 4f,
                    bottom = size.height
                )
            )
        }
        return Outline.Generic(path = path)
    }
}

Image(
    painter = painterResource(id = R.drawable.dog),
    contentDescription = stringResource(id = R.string.dog_content_description),
    contentScale = ContentScale.Crop,
    modifier = Modifier
        .size(200.dp)
        .clip(SquashedOval())
)

A square image of a dog clipped into a custom, oval shape.
Figure 3. Clipping an image with a custom path shape.

Add a border to an Image composable

A common operation is to combine the Modifier.border() with Modifier.clip() to create a border around an image:

val borderWidth = 4.dp
Image(
    painter = painterResource(id = R.drawable.dog),
    contentDescription = stringResource(id = R.string.dog_content_description),
    contentScale = ContentScale.Crop,
    modifier = Modifier
        .size(150.dp)
        .border(
            BorderStroke(borderWidth, Color.Yellow),
            CircleShape
        )
        .padding(borderWidth)
        .clip(CircleShape)
)

A square image of a dog, clipped into a circle, with a yellow border around the circular shape.
Figure 4. A clipped image with a border around it.

To create a gradient border, you can use the Brush API to draw a rainbow gradient border around the image:

val rainbowColorsBrush = remember {
    Brush.sweepGradient(
        listOf(
            Color(0xFF9575CD),
            Color(0xFFBA68C8),
            Color(0xFFE57373),
            Color(0xFFFFB74D),
            Color(0xFFFFF176),
            Color(0xFFAED581),
            Color(0xFF4DD0E1),
            Color(0xFF9575CD)
        )
    )
}
val borderWidth = 4.dp
Image(
    painter = painterResource(id = R.drawable.dog),
    contentDescription = stringResource(id = R.string.dog_content_description),
    contentScale = ContentScale.Crop,
    modifier = Modifier
        .size(150.dp)
        .border(
            BorderStroke(borderWidth, rainbowColorsBrush),
            CircleShape
        )
        .padding(borderWidth)
        .clip(CircleShape)
)

A circular image of a dog with a rainbow gradient border around the circular shape.
Figure 5. Rainbow gradient circle border.

Set a custom aspect ratio

To transform an image into a custom aspect ratio, use Modifier.aspectRatio(16f/9f) to provide a custom ratio for an image (or any composable).

Image(
    painter = painterResource(id = R.drawable.dog),
    contentDescription = stringResource(id = R.string.dog_content_description),
    modifier = Modifier.aspectRatio(16f / 9f)
)

A square image of a dog, transformed to a 16:9 aspect ratio, making it wider and shorter.
Figure 6. Using Modifier.aspectRatio(16f/9f) on an Image.

Color filter: transform pixel colors of image

The Image composable has a colorFilter parameter that can change the output of individual pixels of your image.

Tint images

Using ColorFilter.tint(color, blendMode) applies a blend mode with the given color onto your Image composable. ColorFilter.tint(color, blendMode) uses BlendMode.SrcIn to tint content, meaning that the color supplied is shown where the image is displayed on screen. This is useful for icons and vectors that need to be themed differently.

Image(
    painter = painterResource(id = R.drawable.baseline_directions_bus_24),
    contentDescription = stringResource(id = R.string.bus_content_description),
    colorFilter = ColorFilter.tint(Color.Yellow)
)

An image of a bus with a yellow tint applied.
Figure 7. ColorFilter.tint applied with BlendMode.SrcIn.

Other BlendModes result in different effects. For example, setting BlendMode.Darken with a Color.Green on an image produces the following result:

Image(
    painter = painterResource(id = R.drawable.dog),
    contentDescription = stringResource(id = R.string.dog_content_description),
    colorFilter = ColorFilter.tint(Color.Green, blendMode = BlendMode.Darken)
)

A a dog with a green tint applied using BlendMode.Darken, resulting in darker green shades.
Figure 8. Color.Green tint with BlendMode.Darken.

See the BlendMode reference documentation for more information on the different blend modes available.

Apply an Image filter with color matrix

Transform your image using the color matrix ColorFilter option. For example, to apply a black and white filter onto your images you could use the ColorMatrix and set the saturation to 0f.

Image(
    painter = painterResource(id = R.drawable.dog),
    contentDescription = stringResource(id = R.string.dog_content_description),
    colorFilter = ColorFilter.colorMatrix(ColorMatrix().apply { setToSaturation(0f) })
)

A dog with a black and white filter applied, removing all color saturation.
Figure 9. Color matrix with saturation 0 (black and white image).

Adjust contrast or brightness of an Image composable

To change the contrast and brightness of an image, you can use the ColorMatrix to change the values:

val contrast = 2f // 0f..10f (1 should be default)
val brightness = -180f // -255f..255f (0 should be default)
val colorMatrix = floatArrayOf(
    contrast, 0f, 0f, 0f, brightness,
    0f, contrast, 0f, 0f, brightness,
    0f, 0f, contrast, 0f, brightness,
    0f, 0f, 0f, 1f, 0f
)
Image(
    painter = painterResource(id = R.drawable.dog),
    contentDescription = stringResource(id = R.string.dog_content_description),
    colorFilter = ColorFilter.colorMatrix(ColorMatrix(colorMatrix))
)

A dog with increased brightness and contrast, making it appear more vivid.
Figure 10. Adjusted image brightness and contrast using ColorMatrix.

Invert colors of an Image composable

To invert the colors of an image, set the ColorMatrix to invert the colors:

val colorMatrix = floatArrayOf(
    -1f, 0f, 0f, 0f, 255f,
    0f, -1f, 0f, 0f, 255f,
    0f, 0f, -1f, 0f, 255f,
    0f, 0f, 0f, 1f, 0f
)
Image(
    painter = painterResource(id = R.drawable.dog),
    contentDescription = stringResource(id = R.string.dog_content_description),
    colorFilter = ColorFilter.colorMatrix(ColorMatrix(colorMatrix))
)

A dog with its colors inverted, showing a negative-like effect.
Figure 11. Inverted colors on image.

Blur an Image composable

To blur an image, use Modifier.blur(), supplying the radiusX and radiusY, which specifies the blur radius in the horizontal and vertical direction respectively.

Image(
    painter = painterResource(id = R.drawable.dog),
    contentDescription = stringResource(id = R.string.dog_content_description),
    contentScale = ContentScale.Crop,
    modifier = Modifier
        .size(150.dp)
        .blur(
            radiusX = 10.dp,
            radiusY = 10.dp,
            edgeTreatment = BlurredEdgeTreatment(RoundedCornerShape(8.dp))
        )
)

A dog with a strong blur effect applied, making it appear indistinct and out of focus.
Figure 12. BlurEffect applied to an image.

When blurring Images, it is recommended to use BlurredEdgeTreatment(Shape), instead of BlurredEdgeTreatment.Unbounded, as the latter is used for blurring of arbitrary renderings that are expected to render outside the bounds of the original content. For images, it is likely that they won't render outside the bounds of the content, whereas blurring a rounded rectangle may require this distinction.

For example, if we set the BlurredEdgeTreatment to Unbounded on the preceding image, the edges of the image appear blurred instead of sharp:

Image(
    painter = painterResource(id = R.drawable.dog),
    contentDescription = stringResource(id = R.string.dog_content_description),
    contentScale = ContentScale.Crop,
    modifier = Modifier
        .size(150.dp)
        .blur(
            radiusX = 10.dp,
            radiusY = 10.dp,
            edgeTreatment = BlurredEdgeTreatment.Unbounded
        )
        .clip(RoundedCornerShape(8.dp))
)

A blurred image of a dog, where the blur extends beyond the original image boundaries, making the edges fuzzy.
Figure 13. BlurEdgeTreatment.Unbounded.