Advanced anti-aliasing nodes for videos and images. VideoTAADLAA combines TAA with a DLAA-inspired pipeline for sharp, stable results, while VideoAdaptiveAA targets specific aliasing-prone regions. The system utilizes temporal accumulation, jittered sampling, and CNN-based refinement for superior edge quality.

Important

Disclaimer This node does not use NVIDIA's closed-source SDKs or native DLAA/DLSS binaries. Instead, it is a custom adaptation of Temporal and Spatial Anti-Aliasing techniques commonly found in modern game engines, rebuilt entirely from scratch using PyTorch tensor architecture specifically for ComfyUI video post-processing.

Preview

TAA + DLAA Anti-Aliasing	Adaptive Anti-Aliasing

Features

Temporal Anti-Aliasing (TAA): Blends frame history with motion data to eliminate flicker and sub-pixel jitter.
Neural Reconstruction (MSE 0.000023): DLAA-inspired model for near-lossless, high-fidelity edge restoration.
Ghosting-Free Tech: Motion-aware suppression logic to minimize trailing artifacts in high-speed sequences.
4-Offset Jitter: Deterministic sub-pixel sampling for sharper and more consistent reconstruction stability.
Edge-Preserving Clarity: Sobel-based detection preserves fine textures while eliminating aliasing artifacts.
Frequency-Aware Refinement: Frequency and Perceptual loss optimization for crystal-clear, blur-free results.

Comparison

AdaptiveAA Comparison

TAA + DLAA Comparison

🚀 How to use

Test the effect by increasing dlaa_strength to 0.8 or 1.0.

Parameters

taa_strength: Controls the impact of past frames. (Default: 0.6 | Fast Motion: 0.4 | Static: 0.9)
taa_alpha: Determines how quickly history updates. (Default: 0.1 | Aggressive AA: 0.05 | Ghost-free: 0.4)
motion_sensitivity: Threshold to disable TAA in moving areas. (Default: 0.08 | Flicker-free: 0.0 | High-speed: 0.2)
jitter_scale: Amount of sub-pixel shifting for reconstruction. (Default: 0.2 | Off: 0.0 | Maximum: 0.5)
dlaa_strength: Power of the neural reconstruction model. (Default: 0.8 | Subtle: 0.4 | Aggressive: 1.0)
edge_threshold: Sensitivity of edge detection. (Default: 0.15 | Hard edges: 0.3 | Fine details: 0.05)
blur_radius: Radius of pre-AA edge softening. (Default: 1 | Sharp: 0 | Softer: 2)
reset_history: Clears temporal memory. (Default: False | Enable on scene cuts: True)

Installation

Recommended: Install via ComfyUI-Manager.

Manual installation:

cd ComfyUI/custom_nodes/
git clone https://github.com/MSXYZ-GenAI/comfyui-msxyz.git

License

This project is licensed under the MIT License - see the LICENSE file for details.