E-MD3C: Taming Masked Diffusion Transformers for Efficient Zero-Shot Object Customization

• URL: http://arxiv.org/abs/2502.09164v1
• Date: Thu, 13 Feb 2025 10:48:11 GMT
• Title: E-MD3C: Taming Masked Diffusion Transformers for Efficient Zero-Shot Object Customization
• Authors: Trung X. Pham, Zhang Kang, Ji Woo Hong, Xuran Zheng, Chang D. Yoo,
• Abstract summary: E-MD3C is a highly efficient framework for zero-shot object image customization. Unlike prior works reliant on resource-intensive Unet architectures, our approach employs lightweight masked diffusion transformers. E-MD3C outperforms the existing approach on the VITON-HD dataset across metrics such as PSNR, FID, SSIM, and LPIPS.
• Score: 20.441652320245975
• License:
• Abstract: We propose E-MD3C ($\underline{E}$fficient $\underline{M}$asked $\underline{D}$iffusion Transformer with Disentangled $\underline{C}$onditions and $\underline{C}$ompact $\underline{C}$ollector), a highly efficient framework for zero-shot object image customization. Unlike prior works reliant on resource-intensive Unet architectures, our approach employs lightweight masked diffusion transformers operating on latent patches, offering significantly improved computational efficiency. The framework integrates three core components: (1) an efficient masked diffusion transformer for processing autoencoder latents, (2) a disentangled condition design that ensures compactness while preserving background alignment and fine details, and (3) a learnable Conditions Collector that consolidates multiple inputs into a compact representation for efficient denoising and learning. E-MD3C outperforms the existing approach on the VITON-HD dataset across metrics such as PSNR, FID, SSIM, and LPIPS, demonstrating clear advantages in parameters, memory efficiency, and inference speed. With only $\frac{1}{4}$ of the parameters, our Transformer-based 468M model delivers $2.5\times$ faster inference and uses $\frac{2}{3}$ of the GPU memory compared to an 1720M Unet-based latent diffusion model.

Related papers

• Mixture of Hidden-Dimensions Transformer [50.40325486463241]
  We study hidden dimension sparsity and observe that trained Transformers utilize only a small fraction of token dimensions. We propose MoHD (Mixture of Hidden Dimensions), a sparse conditional activation architecture. It achieves 1.7% higher performance with 50% fewer activation parameters and 3.7% higher performance with a 3x parameter expansion at constant activation cost.
  arXiv  Detail & Related papers  (2024-12-07T13:15:22Z)
• CARE Transformer: Mobile-Friendly Linear Visual Transformer via Decoupled Dual Interaction [77.8576094863446]
  We propose a new detextbfCoupled dutextbfAl-interactive lineatextbfR atttextbfEntion (CARE) mechanism. We first propose an asymmetrical feature decoupling strategy that asymmetrically decouples the learning process for local inductive bias and long-range dependencies. By adopting a decoupled learning way and fully exploiting complementarity across features, our method can achieve both high efficiency and accuracy.
  arXiv  Detail & Related papers  (2024-11-25T07:56:13Z)
• Dynamic Diffusion Transformer [67.13876021157887]
  Diffusion Transformer (DiT) has demonstrated superior performance but suffers from substantial computational costs. We propose Dynamic Diffusion Transformer (DyDiT), an architecture that dynamically adjusts its computation along both timestep and spatial dimensions during generation. With 3% additional fine-tuning, our method reduces the FLOPs of DiT-XL by 51%, accelerates generation by 1.73, and achieves a competitive FID score of 2.07 on ImageNet.
  arXiv  Detail & Related papers  (2024-10-04T14:14:28Z)
• Stretching Each Dollar: Diffusion Training from Scratch on a Micro-Budget [53.311109531586844]
  We demonstrate very low-cost training of large-scale T2I diffusion transformer models. We train a 1.16 billion parameter sparse transformer with only $1,890 economical cost and achieve a 12.7 FID in zero-shot generation. We aim to release our end-to-end training pipeline to further democratize the training of large-scale diffusion models on micro-budgets.
  arXiv  Detail & Related papers  (2024-07-22T17:23:28Z)
• Cross-view Masked Diffusion Transformers for Person Image Synthesis [21.242398582282522]
  We present X-MDPT, a novel diffusion model designed for pose-guided human image generation. X-MDPT distinguishes itself by employing masked diffusion transformers that operate on latent patches. Our model outperforms state-of-the-art approaches on the DeepFashion dataset.
  arXiv  Detail & Related papers  (2024-02-02T15:57:13Z)
• Hourglass Tokenizer for Efficient Transformer-Based 3D Human Pose Estimation [73.31524865643709]
  We present a plug-and-play pruning-and-recovering framework, called Hourglass Tokenizer (HoT), for efficient transformer-based 3D pose estimation from videos. Our HoDT begins with pruning pose tokens of redundant frames and ends with recovering full-length tokens, resulting in a few pose tokens in the intermediate transformer blocks. Our method can achieve both high efficiency and estimation accuracy compared to the original VPT models.
  arXiv  Detail & Related papers  (2023-11-20T18:59:51Z)
• UNETR++: Delving into Efficient and Accurate 3D Medical Image Segmentation [93.88170217725805]
  We propose a 3D medical image segmentation approach, named UNETR++, that offers both high-quality segmentation masks as well as efficiency in terms of parameters, compute cost, and inference speed. The core of our design is the introduction of a novel efficient paired attention (EPA) block that efficiently learns spatial and channel-wise discriminative features. Our evaluations on five benchmarks, Synapse, BTCV, ACDC, BRaTs, and Decathlon-Lung, reveal the effectiveness of our contributions in terms of both efficiency and accuracy.
  arXiv  Detail & Related papers  (2022-12-08T18:59:57Z)
• Fcaformer: Forward Cross Attention in Hybrid Vision Transformer [29.09883780571206]
  We propose forward cross attention for hybrid vision transformer (FcaFormer) Our FcaFormer achieves 83.1% top-1 accuracy on Imagenet with only 16.3 million parameters and about 3.6 billion MACs. This saves almost half of the parameters and a few computational costs while achieving 0.7% higher accuracy compared to distilled EfficientFormer.
  arXiv  Detail & Related papers  (2022-11-14T08:43:44Z)
• EcoFormer: Energy-Saving Attention with Linear Complexity [40.002608785252164]
  Transformer is a transformative framework that models sequential data. We propose a new binarization paradigm customized to high-dimensional softmax attention. We show that EcoFormer consistently achieves comparable performance with standard attentions.
  arXiv  Detail & Related papers  (2022-09-19T13:28:32Z)
• MVSTER: Epipolar Transformer for Efficient Multi-View Stereo [26.640495084316925]
  Learning-based Multi-View Stereo methods warp source images into 3D volumes, which are fused as a cost volume to be regularized by subsequent networks. Previous methods utilize extra networks to learn 2D information as fusing cues, underusing 3D spatial correlations and bringing additional computation costs. We present MVSTER, which leverages the proposed epipolar Transformer to learn both 2D semantics and 3D spatial associations efficiently.
  arXiv  Detail & Related papers  (2022-04-15T06:47:57Z)
• AFTer-UNet: Axial Fusion Transformer UNet for Medical Image Segmentation [19.53151547706724]
  transformer-based models have drawn attention to exploring these techniques in medical image segmentation. We propose Axial Fusion Transformer UNet (AFTer-UNet), which takes both advantages of convolutional layers' capability of extracting detailed features and transformers' strength on long sequence modeling. It has fewer parameters and takes less GPU memory to train than the previous transformer-based models.
  arXiv  Detail & Related papers  (2021-10-20T06:47:28Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.