NEXTLIFE — 3D Objects and Distortions OVERVIEW NEXTLIFE is a dataset of 3D object models, including their materials, textures, and distorted variants. It supports research into how distortions affect object recognition and visual security. The current selection contains 119 retained objects and 3,332 distorted variants, corresponding to 28 profiles per object. The transformations include mesh simplification and quantization, texture resizing, compression and blurring, as well as encryption and obscuration. The mesh “data hiding” profile currently uses an encryption-based proxy. DOWNLOAD CONTENTS The assets are provided as three separate ZIP archives, split into volumes of at most 4 GiB, following the .zip.001, .zip.002, etc. naming convention: - NEXTLIFE_20260930_source.zip.001: the 119 retained reference objects (one volume). - NEXTLIFE_20260930_distorted.zip.001, .zip.002, .zip.003, and .zip.004: their 3,332 distorted variants (four volumes; each filename starts with NEXTLIFE_20260930_distorted). - NEXTLIFE_20260930_scenes.zip.001: the five active experimental environments and their required resources (one volume). There are six archive volumes in total, occupying approximately 13.8 GiB. All three archives have passed 7-Zip integrity checks and file-inventory verification. Download every volume belonging to each archive. Keep the original filenames and place all volumes together in one download folder. Start extraction from .zip.001 only; do not extract each volume separately or rename it to .zip. Use 7-Zip to handle these split archives. The archives preserve the repository folder structure and extract to: - Objects/Originals/: reference objects, materials, and textures. - Objects/Distorted/MeshVariants/: transformed meshes. - Objects/Distorted/TextureVariants/: transformed textures. - Objects/Distorted/CombinedVariants/: manifests linking meshes and textures. - Objects/Distorted/dataset_info.json: the validated object selection and dataset identifier. - Scenes/: experimental environments, provided in the scenes archive. Keep all filenames and the complete folder structure unchanged. A variant manifest references other files; it is not a standalone 3D object. Objects/Distorted/dataset_info.json preserves the validated selection and the stable dataset identifier. Excluded reference objects are not included in this distribution. The dataset_info.json supplied beside the archives describes the download volumes and their sizes. SHA256SUMS.txt contains SHA-256 checksums for the archive volumes. These delivery files are separate from Objects/Distorted/dataset_info.json, which is included inside the distorted archive. CODE AND TOOLS The 3D assets are distributed separately from the code. The viewers, scripts, configuration files, and metadata needed to run the experiments are available on GitHub: https://github.com/Kaldrass/Dataset_NEXTLIFE These tools allow users to explore objects, compare their distorted variants, and run an object recognition and visual security experiment. WINDOWS INSTALLATION 1. Install Git, Python 3.12, and 7-Zip. 2. Open PowerShell in a working directory and run the following commands, one line at a time: git clone https://github.com/Kaldrass/Dataset_NEXTLIFE.git cd Dataset_NEXTLIFE python -m venv .venv .\.venv\Scripts\python.exe -m pip install -r requirements.txt 3. Download all archive volumes into one folder, for example D:\Downloads\NEXTLIFE. In the PowerShell window opened at the root of the cloned repository, run the following commands. Change the download path to match your computer: $assets = "D:\Downloads\NEXTLIFE" & "$env:ProgramFiles\7-Zip\7z.exe" x "$assets\NEXTLIFE_20260930_source.zip.001" -o. & "$env:ProgramFiles\7-Zip\7z.exe" x "$assets\NEXTLIFE_20260930_distorted.zip.001" -o. & "$env:ProgramFiles\7-Zip\7z.exe" x "$assets\NEXTLIFE_20260930_scenes.zip.001" -o. These commands assume 7-Zip is installed in its standard Windows location. Each command automatically reads the remaining volumes. Extract into a fresh clone to avoid mixing asset versions. The archives create Objects/Originals/, Objects/Distorted/, and Scenes/ directly beside metadata.json and ExperimentSecurity/. Do not extract into an additional Objects/ folder. Alternatively, in the 7-Zip application, open the .zip.001 file for each archive, choose Extract, and select the root of the cloned repository as the destination. To check a downloaded volume in PowerShell, run the following command and compare the result with its entry in SHA256SUMS.txt: Get-FileHash "$assets\NEXTLIFE_20260930_source.zip.001" -Algorithm SHA256 Repeat for the other volumes. Allow enough disk space for both the downloaded archives and approximately 50 GiB of extracted assets. On Linux or macOS, with the 7-Zip command-line tool installed, run these commands from the repository root, adapting the download path: 7zz x /path/to/downloads/NEXTLIFE_20260930_source.zip.001 -o. 7zz x /path/to/downloads/NEXTLIFE_20260930_distorted.zip.001 -o. 7zz x /path/to/downloads/NEXTLIFE_20260930_scenes.zip.001 -o. For the Python commands in this guide on Linux/macOS, create the environment with python3 -m venv .venv and use .venv/bin/python instead of .\.venv\Scripts\python.exe. Use forward slashes in script paths. The distorted variants have already been generated. There is no need to regenerate them to use the dataset. EXPLORATION AND DEMONSTRATION SESSION From the repository root, build the catalog: .\.venv\Scripts\python.exe ExperimentSecurity\build_dataset_catalog.py Prepare a demonstration session with up to 20 trials and up to 5 distinct objects: .\.venv\Scripts\python.exe ExperimentSecurity\build_recognition_experiment.py --max-objects 5 --max-trials 20 --max-faces 0 --choices 6 --seed 20260622 This command replaces recognition_trials.json. Back up any existing trial set before changing it. This small session is intended to check that the tools work; it is not a balanced experimental design. Start the local server: .\.venv\Scripts\python.exe -m http.server 8015 --bind 127.0.0.1 Keep the terminal open, then open one of the following addresses in Chrome. These addresses work on the computer running the local server. Dataset explorer: http://127.0.0.1:8015/ExperimentSecurity/dataset_explorer.html Participant experiment: http://127.0.0.1:8015/ExperimentSecurity/recognition_viewer.html The viewers load Three.js from the Internet. If the scenes are missing, a fallback environment is used: obtain the intended scenes before running an actual study. To stop the server, press Ctrl + C in the terminal. RESULTS AND DOCUMENTATION During the experiment, participants choose an object label and a visual security level, then click “Valider” (Validate). After validating the final trial, export the responses using the JSON or CSV buttons. Name each export using an anonymous participant identifier. Responses are stored in the browser and are not automatically sent to the server: export them before using Reset for the next participant. Analysis commands and detailed usage instructions are available here: https://github.com/Kaldrass/Dataset_NEXTLIFE/blob/main/ExperimentSecurity/README.md ExperimentDSIS/ contains a separate historical experiment. The Old/ archives, temporary files, and earlier generations are not needed to use the active dataset assets.