thehardwarestore.ai

The code answers the research question of how to customize large text-to-image diffusion models with minimal data and minimal parameter updates, and it compares against DreamBooth and Textual Inversion. It includes reusable components for few-shot personalization, delta checkpoint handling, lightweight model compression, multi-concept weight composition, and CLIP/DINO evaluation, although the codebase shows research-level roughness: duplicated CompVis/diffusers paths, hardcoded paths, and several known bugs in conversion and composition utilities.

Show-o and Show-o2 are research repositories for unified multimodal understanding and generation. They implement transformer-based models (1.3B, 1.5B, 7B parameters) that can perform text-to-image generation, image/video captioning, visual question answering, inpainting/extrapolation, and mixed-modality generation (interleaved text and images). The codebase is organized around two major model versions: Show-o uses discrete tokens with MAGVITv2 VAE and Phi-1.5 LLM, while Show-o2 moves to continuous latent space with WanVAE (3D causal VAE) and Qwen2.5 LLM. Training follows a multi-stage pipeline (pretraining, instruction tuning) with novel techniques like flow matching (via a modular `transport` package), dual-path vision encoders (SigLIP + patch embedding), and omni-attention masks that combine causal and bidirectional attention for multimodal sequences. The repository includes distributed training configs (DeepSpeed, multi-node), evaluation scripts for GenEval and DPG-Bench, and comprehensive YAML configuration files. This is primarily a data_research codebase: it produces research artifacts (models, papers at ICLR 2025 and NeurIPS 2025) rather than a production application. The code exhibits good modularity for a research project, with reusable components like the transport module, omni-attention masks, mixed dataloader, and attention masking utilities. However, it suffers from typical research-code issues: hardcoded internal infrastructure paths, reliance on non-public data pipelines (e.g., internal packages for RefinedWeb), and some code duplication between Show-o and Show-o2. Reproducibility is moderate—dependencies are pinned and configs are detailed, but many datasets are large-scale web sources that require manual setup, and the training scripts assume specific hardware (multi-node A100 clusters). The project is actively maintained (commits in early 2026) with a roadmap for future work, including mixed-modal generation and video support.

UniTEX is a research-focused repository implementing a two-stage framework for high-fidelity 3D texture generation from arbitrary 3D shapes. Stage 1 uses LoRA-adapted Diffusion Transformers for multi-view texture synthesis, while Stage 2 employs a transformer-based Large Texturing Model (LTM) to predict Texture Functions in unified 3D space, bypassing UV mapping limitations. The repository integrates super-resolution (TSD_SR) for detail enhancement and provides tools for mesh processing, rendering, and texture reprojection. It includes training scripts, model architectures, and utility libraries, indicating its primary purpose is to support experimentation and reproduction of the CVPR 2026 paper's methodology rather than serving as a standalone end-user application.

This repository implements Geometry Forcing, a research framework for video generation with geometric consistency. The core contribution marries video diffusion models (DFoT - Diffusion Forcing Transformer) with 3D representation learning (VGGT) to produce temporally consistent videos conditioned on camera poses. It includes full training pipelines (Hydra + PyTorch Lightning), a suite of video metrics (FVD, FID, LPIPS, VBench), and evaluation scripts for reprojection/revisit error using DROID-SLAM. The codebase is structured as a research template with reusable base classes for algorithms, data modules, experiments, and metric registries. Raw datasets (RealEstate10K, Kinetics-600, Minecraft) are downloaded and preprocessed into video clips with associated camera parameters. Transformations include VAE encoding for latent diffusion, augmentation (flip, reverse), and sliding-window clip extraction. The primary research question is whether video diffusion can be constrained by explicit 3D geometry to reduce inconsistencies in generated videos, measured by geometry-aware metrics. The repo also demonstrates composable sampling via History Guidance and an alignment loss that projects diffusion latents into VGGT feature space. While the code is well-engineered for research (modular design, config-driven, metric registry), it carries typical research code baggage: hardcoded paths, vendored third-party code (DROID-SLAM, VGGT), potential dependency pinning conflicts, and some dead/commented code. Reproducibility is moderate: dependencies are pinned but no environment definitions beyond requirements.txt, and dataset downloads rely on fragile YouTube scraping.

This repository is a curated, annotated bibliography (awesome list) of research papers on style transfer using diffusion models. It contains no executable code, datasets, or pipelines. The value lies in taxonomizing papers into categories such as Image-Encoder Based, Swap Features in Attention Layer, Optimization-Based, LoRA-Based, Video Synthesis, 3D Generation, and Style Datasets. Each paper is listed with title, publication venue, date, and a link to its GitHub repository if available. The README also includes a brief perspective from the maintainers on the taxonomy's motivation. While the list itself is useful for researchers and practitioners in computer vision and generative models, the repository is not a functional software project and provides no reusable code or data assets.