Canonical: https://instain.ru/en/ai/apisr/ Language: en Zhilu Wang and co-authors model # APISR. Upscale animation frames and illustrations 4.25 / 5 [Directory score](https://instain.ru/en/ai/apisr/#editorial-rating) [Official website ](https://github.com/Kiteretsu77/APISR) The model of increasing resolution for painted frames. The project offers several processing scales and examples of running on images and videos. ![APISR. APISR repository with scales and examples of increasing the framed frames.](https://instain.ru/assets/images/apisr-screen-70b972b65e7c-1200.webp) APISR repository with scales and examples of increasing the framed frames. [Open the source ](https://github.com/Kiteretsu77/APISR) ## Why a painted frame is a separate model APISR is designed to increase low-quality painted frames. Its authors took into account the properties of materials from the production of animation, where a sharp thin line often passes near a flat fill. If the image has survived recompression, the line can disintegrate into small steps. The model offers a recovery that needs to be evaluated on the material of a particular drawing. Sometimes a sharper version looks foreign next to the rest of the footage. ## Check on your own short fragment Take a training video with a moving character and a simple background. Start with a single frame, which shows the eyelashes, the edge of the sleeve and a small inscription. Compare the two dimensions of increase. If the inscription has changed or the strokes are missing, do not transfer the result to the whole series. After that, process several neighboring frames and view them in a row to notice a flicker on the moving contour. Keep the trial files separate. When a larger pattern is shown in the overall layout, the readability of the line is more important than the expression of each new detail. ## Where to choose weight and mode The official repository contains a model catalog, instructions for processing images and video, and examples at different scales. Record the chosen weights and processing settings for your workflow. Check that you have permission to transform and share the source materials. Publishing an enlarged copy may involve different requirements from evaluating it on your computer. Show the original and processed frames side by side for review. ## Suitable tasks - Prepare an enlarged illustration for the layout - Compare short animation processing Features ## Strengths - The model is focused on painted contours and frames - Different increase rates are available Things to consider ## Limitations - The restored lines may differ from the original - Individual frames need to be checked for flickering ## Selection and conditions Free access Yes Trial period Not confirmed No sign-up required Not confirmed Russian interface Not confirmed Local installation Yes API Not confirmed “Not confirmed” means there is not enough information. The feature may still be available; check the developer’s terms. Input formats Image, Video Result Image, Video ### Terms Authors publish code, model weights and instructions. For a sample on your machine, check for Python dependencies and image or video processing mode. From task to first result ## Practical example Task Prepare an enlarged illustration for the layout What to prepare A copy of your own drawing or animation fragment with compression traces. ### Task brief or prompt Increase the copy at the right scale and check if thin lines and flat fills have changed. ### Expected result Enlarged version for viewing with marked artifacts. ### How to check the result - Examine the strokes, inscriptions and transitions of color. - For the roller, check the stability of the contours from frame to frame. An editorial teaching example of a task brief, not the result of hands-on testing of the service. Check that the required mode is available, that you have rights to the source material and that the intended use of the result is permitted. Check details ## Official sources - [Official APISR code, weights and examples ](https://github.com/Kiteretsu77/APISR) - [The work of APISR authors on increasing the drawn graphics ](https://arxiv.org/abs/2403.01598) This description is based on the developer’s material. Features, prices and availability may change; check the primary source before use. About the tool Developer Zhilu Wang and co-authors Category [Images](https://instain.ru/en/category/images/) Access Authors publish code, model weights and instructions. For a sample on your machine, check for Python dependencies and image or video processing mode. Payment The project code is published under GPL 3.0. Check the terms of the scales used and the rights to the original animation before distributing the processing. API No separate API specified [Developer terms ](https://github.com/Kiteretsu77/APISR) Directory score 4.25 / 5 Features 4.30 / 5 Workflows 4.10 / 5 Ecosystem 4.20 / 5 Documentation 4.40 / 5 Scores are an editorial guide to the directory. They are not user reviews or the results of personal testing. [See the methodology ](https://instain.ru/en/methodology/)