2024-09-02 更新
Enabling Local Editing in Diffusion Models by Joint and Individual Component Analysis
Authors:Theodoros Kouzelis, Manos Plitsis, Mihalis A. Nikolaou, Yannis Panagakis
Recent advances in Diffusion Models (DMs) have led to significant progress in visual synthesis and editing tasks, establishing them as a strong competitor to Generative Adversarial Networks (GANs). However, the latent space of DMs is not as well understood as that of GANs. Recent research has focused on unsupervised semantic discovery in the latent space of DMs by leveraging the bottleneck layer of the denoising network, which has been shown to exhibit properties of a semantic latent space. However, these approaches are limited to discovering global attributes. In this paper we address, the challenge of local image manipulation in DMs and introduce an unsupervised method to factorize the latent semantics learned by the denoising network of pre-trained DMs. Given an arbitrary image and defined regions of interest, we utilize the Jacobian of the denoising network to establish a relation between the regions of interest and their corresponding subspaces in the latent space. Furthermore, we disentangle the joint and individual components of these subspaces to identify latent directions that enable local image manipulation. Once discovered, these directions can be applied to different images to produce semantically consistent edits, making our method suitable for practical applications. Experimental results on various datasets demonstrate that our method can produce semantic edits that are more localized and have better fidelity compared to the state-of-the-art.
PDF Code available here: https://zelaki.github.io/localdiff/
点此查看论文截图
Revising Multimodal VAEs with Diffusion Decoders
Authors:Daniel Wesego, Amirmohammad Rooshenas
Multimodal VAEs often struggle with generating high-quality outputs, a challenge that extends beyond the inherent limitations of the VAE framework. The core issue lies in the restricted joint representation of the latent space, particularly when complex modalities like images are involved. Feedforward decoders, commonly used for these intricate modalities, inadvertently constrain the joint latent space, leading to a degradation in the quality of the other modalities as well. Although recent studies have shown improvement by introducing modality-specific representations, the issue remains significant. In this work, we demonstrate that incorporating a flexible diffusion decoder specifically for the image modality not only enhances the generation quality of the images but also positively impacts the performance of the other modalities that rely on feedforward decoders. This approach addresses the limitations imposed by conventional joint representations and opens up new possibilities for improving multimodal generation tasks using the multimodal VAE framework. Our model provides state-of-the-art results compared to other multimodal VAEs in different datasets with higher coherence and superior quality in the generated modalities
PDF
点此查看论文截图
Contrastive Learning with Synthetic Positives
Authors:Dewen Zeng, Yawen Wu, Xinrong Hu, Xiaowei Xu, Yiyu Shi
Contrastive learning with the nearest neighbor has proved to be one of the most efficient self-supervised learning (SSL) techniques by utilizing the similarity of multiple instances within the same class. However, its efficacy is constrained as the nearest neighbor algorithm primarily identifies easy'' positive pairs, where the representations are already closely located in the embedding space. In this paper, we introduce a novel approach called Contrastive Learning with Synthetic Positives (CLSP) that utilizes synthetic images, generated by an unconditional diffusion model, as the additional positives to help the model learn from diverse positives. Through feature interpolation in the diffusion model sampling process, we generate images with distinct backgrounds yet similar semantic content to the anchor image. These images are consideredhard’’ positives for the anchor image, and when included as supplementary positives in the contrastive loss, they contribute to a performance improvement of over 2\% and 1\% in linear evaluation compared to the previous NNCLR and All4One methods across multiple benchmark datasets such as CIFAR10, achieving state-of-the-art methods. On transfer learning benchmarks, CLSP outperforms existing SSL frameworks on 6 out of 8 downstream datasets. We believe CLSP establishes a valuable baseline for future SSL studies incorporating synthetic data in the training process.
PDF 8 pages, conference
点此查看论文截图
Text-to-Image Generation Via Energy-Based CLIP
Authors:Roy Ganz, Michael Elad
Joint Energy Models (JEMs), while drawing significant research attention, have not been successfully scaled to real-world, high-resolution datasets. We present EB-CLIP, a novel approach extending JEMs to the multimodal vision-language domain using CLIP, integrating both generative and discriminative objectives. For the generative objective, we introduce an image-text joint-energy function based on Cosine similarity in the CLIP space, training CLIP to assign low energy to real image-caption pairs and high energy otherwise. For the discriminative objective, we employ contrastive adversarial loss, extending the adversarial training objective to the multimodal domain. EB-CLIP not only generates realistic images from text but also achieves competitive results on the compositionality benchmark, outperforming leading methods with fewer parameters. Additionally, we demonstrate the superior guidance capability of EB-CLIP by enhancing CLIP-based generative frameworks and converting unconditional diffusion models to text-based ones. Lastly, we show that EB-CLIP can serve as a more robust evaluation metric for text-to-image generative tasks than CLIP.
PDF
点此查看论文截图
VQ4DiT: Efficient Post-Training Vector Quantization for Diffusion Transformers
Authors:Juncan Deng, Shuaiting Li, Zeyu Wang, Hong Gu, Kedong Xu, Kejie Huang
The Diffusion Transformers Models (DiTs) have transitioned the network architecture from traditional UNets to transformers, demonstrating exceptional capabilities in image generation. Although DiTs have been widely applied to high-definition video generation tasks, their large parameter size hinders inference on edge devices. Vector quantization (VQ) can decompose model weight into a codebook and assignments, allowing extreme weight quantization and significantly reducing memory usage. In this paper, we propose VQ4DiT, a fast post-training vector quantization method for DiTs. We found that traditional VQ methods calibrate only the codebook without calibrating the assignments. This leads to weight sub-vectors being incorrectly assigned to the same assignment, providing inconsistent gradients to the codebook and resulting in a suboptimal result. To address this challenge, VQ4DiT calculates the candidate assignment set for each weight sub-vector based on Euclidean distance and reconstructs the sub-vector based on the weighted average. Then, using the zero-data and block-wise calibration method, the optimal assignment from the set is efficiently selected while calibrating the codebook. VQ4DiT quantizes a DiT XL/2 model on a single NVIDIA A100 GPU within 20 minutes to 5 hours depending on the different quantization settings. Experiments show that VQ4DiT establishes a new state-of-the-art in model size and performance trade-offs, quantizing weights to 2-bit precision while retaining acceptable image generation quality.
PDF 11 pages, 6 figures
点此查看论文截图
CinePreGen: Camera Controllable Video Previsualization via Engine-powered Diffusion
Authors:Yiran Chen, Anyi Rao, Xuekun Jiang, Shishi Xiao, Ruiqing Ma, Zeyu Wang, Hui Xiong, Bo Dai
With advancements in video generative AI models (e.g., SORA), creators are increasingly using these techniques to enhance video previsualization. However, they face challenges with incomplete and mismatched AI workflows. Existing methods mainly rely on text descriptions and struggle with camera placement, a key component of previsualization. To address these issues, we introduce CinePreGen, a visual previsualization system enhanced with engine-powered diffusion. It features a novel camera and storyboard interface that offers dynamic control, from global to local camera adjustments. This is combined with a user-friendly AI rendering workflow, which aims to achieve consistent results through multi-masked IP-Adapter and engine simulation guidelines. In our comprehensive evaluation study, we demonstrate that our system reduces development viscosity (i.e., the complexity and challenges in the development process), meets users’ needs for extensive control and iteration in the design process, and outperforms other AI video production workflows in cinematic camera movement, as shown by our experiments and a within-subjects user study. With its intuitive camera controls and realistic rendering of camera motion, CinePreGen shows great potential for improving video production for both individual creators and industry professionals.
PDF
点此查看论文截图