About

Modifying colors in an animated GIF requires decoding every frame from the GIF89a binary format, decompressing LZW-encoded pixel data, transforming each pixel in a chosen color space, then re-encoding the entire sequence. A single mishandled disposal method or local color table will corrupt every subsequent frame. This tool parses the raw binary structure, extracts frames with correct timing and disposal, applies color transformations in HSL space for perceptually uniform shifts, and re-encodes with proper GCT palettes. Tolerance-based color replacement uses Euclidean distance d = √ΔR² + ΔG² + ΔB² in RGB space to catch neighboring shades. Note: heavily dithered GIFs with limited 256-color palettes may show banding after aggressive hue shifts. Transparency frames are preserved but disposal method “restore to background” may produce artifacts on GIFs that rely on frame stacking.

Formulas

Hue shifting operates in HSL color space. Each pixel's RGB values are converted, the hue channel is rotated, then converted back.

H_new = (H_original + ΔH) mod 360°

For targeted color replacement, the Euclidean distance in RGB space determines pixel eligibility:

d = √(R₁ − R₂)² + (G₁ − G₂)² + (B₁ − B₂)²

A pixel is replaced when d ≤ T (tolerance). Maximum possible distance is √3 × 255² ≈ 441.7. Grayscale conversion uses the luminance formula per ITU-R BT.709:

L = 0.2126R + 0.7152G + 0.0722B

Where R, G, B are the pixel's red, green, blue channel values (0 - 255). H is hue in degrees (0 - 360). ΔH is the user-specified hue rotation. T is the color match tolerance. d is the Euclidean distance between the target and pixel color.

Reference Data

Color Operation	Color Space	Best For	Limitation
Hue Shift	HSL	Global color theme change	Grays unaffected (saturation ≈ 0)
Saturation Adjust	HSL	Vivid ↔ muted tones	Clipping at 0 or 100%
Brightness Adjust	HSL	Lighten or darken	Destroys detail at extremes
Color Replace	RGB	Swap specific color	Needs tolerance for anti-aliased edges
Invert	RGB	Negative effect	Transparency unchanged
Grayscale	RGB (luminance)	Remove all color	Irreversible once saved
Sepia	RGB (matrix)	Vintage/warm tone	Shifts blues to brown
Channel Isolate (R)	RGB	Red channel only	Loses all non-red info
Channel Isolate (G)	RGB	Green channel only	Loses all non-green info
Channel Isolate (B)	RGB	Blue channel only	Loses all non-blue info
Threshold	Luminance	Black & white cutoff	Loses gradients entirely
Posterize	RGB	Reduce color levels	Banding artifacts
Temperature Shift	RGB (R/B bias)	Warm/cool tone	Can clip channels
Tint Overlay	RGB blend	Uniform color wash	Reduces contrast

Frequently Asked Questions

Hue shifting operates in HSL color space. Pixels with zero or near-zero saturation (grays, whites, blacks) have no meaningful hue to rotate. Only chromatic pixels are affected. If your GIF is heavily desaturated, increase the saturation slider first, then apply the hue shift.

GIF format supports a maximum of 256 colors per frame. When your color transformation creates gradients that exceed this limit, the encoder must quantize colors, producing visible steps. This is most noticeable with subtle hue shifts on photographic GIFs. Reducing the number of unique colors in your transformation (e.g., using posterize first) can mitigate this.

Tolerance is the maximum Euclidean distance in RGB space between the target color and a pixel for it to be considered a match. A tolerance of 0 matches only exact pixel values. A tolerance of 50 catches similar shades (useful for anti-aliased edges). The maximum meaningful value is approximately 441 (distance from black to white). Values above 100 will typically replace large portions of the image.

GIF transparency is index-based - one palette entry is flagged as transparent. During re-encoding, the tool preserves pixels with alpha = 0 as transparent. However, if your color transformation modifies what was originally the transparent color index, the mapping may shift. The tool handles this by tracking alpha channels separately from color manipulation.

The tool processes GIFs up to 20 MB. Frame count is limited by available browser memory - typically 500+ frames work reliably. Processing time scales linearly with total pixel count (width × height × frames). A 500×500 pixel GIF with 100 frames processes roughly 25 million pixels. Expect 5-15 seconds depending on device performance.

Yes. GIF frames use disposal methods to determine how each frame composites onto the previous one. Method 1 (do not dispose) stacks frames. Method 2 (restore to background) clears to background color. This tool decodes each frame into a full composite canvas before applying color changes, ensuring disposal methods are correctly handled. The output uses disposal method 1 with full frames for maximum compatibility.