Diffusion-based Visual Anagram as Multi-task Learning
- URL: http://arxiv.org/abs/2412.02693v1
- Date: Tue, 03 Dec 2024 18:59:28 GMT
- Title: Diffusion-based Visual Anagram as Multi-task Learning
- Authors: Zhiyuan Xu, Yinhe Chen, Huan-ang Gao, Weiyan Zhao, Guiyu Zhang, Hao Zhao,
- Abstract summary: A visual anagram is an image that changes appearance upon transformation, like flipping or rotation.
We propose a new method to generate true anagrams spanning diverse concepts.
- Score: 9.233197675701025
- License:
- Abstract: Visual anagrams are images that change appearance upon transformation, like flipping or rotation. With the advent of diffusion models, generating such optical illusions can be achieved by averaging noise across multiple views during the reverse denoising process. However, we observe two critical failure modes in this approach: (i) concept segregation, where concepts in different views are independently generated, which can not be considered a true anagram, and (ii) concept domination, where certain concepts overpower others. In this work, we cast the visual anagram generation problem in a multi-task learning setting, where different viewpoint prompts are analogous to different tasks,and derive denoising trajectories that align well across tasks simultaneously. At the core of our designed framework are two newly introduced techniques, where (i) an anti-segregation optimization strategy that promotes overlap in cross-attention maps between different concepts, and (ii) a noise vector balancing method that adaptively adjusts the influence of different tasks. Additionally, we observe that directly averaging noise predictions yields suboptimal performance because statistical properties may not be preserved, prompting us to derive a noise variance rectification method. Extensive qualitative and quantitative experiments demonstrate our method's superior ability to generate visual anagrams spanning diverse concepts.
Related papers
- Enhance Vision-Language Alignment with Noise [59.2608298578913]
We investigate whether the frozen model can be fine-tuned by customized noise.
We propose Positive-incentive Noise (PiNI) which can fine-tune CLIP via injecting noise into both visual and text encoders.
arXiv Detail & Related papers (2024-12-14T12:58:15Z) - Distractors-Immune Representation Learning with Cross-modal Contrastive Regularization for Change Captioning [71.14084801851381]
Change captioning aims to succinctly describe the semantic change between a pair of similar images.
Most existing methods directly capture the difference between them, which risk obtaining error-prone difference features.
We propose a distractors-immune representation learning network that correlates the corresponding channels of two image representations.
arXiv Detail & Related papers (2024-07-16T13:00:33Z) - SemFlow: Binding Semantic Segmentation and Image Synthesis via Rectified Flow [94.90853153808987]
Semantic segmentation and semantic image synthesis are representative tasks in visual perception and generation.
We propose a unified framework (SemFlow) and model them as a pair of reverse problems.
Experiments show that our SemFlow achieves competitive results on semantic segmentation and semantic image synthesis tasks.
arXiv Detail & Related papers (2024-05-30T17:34:40Z) - Visual Anagrams: Generating Multi-View Optical Illusions with Diffusion Models [15.977340635967018]
Multi-view optical illusions are images that change appearance upon a transformation, such as a flip or rotation.
We propose a zero-shot method for obtaining these illusions from off-the-shelf text-to-image diffusion models.
We provide both qualitative and quantitative results demonstrating the effectiveness and flexibility of our method.
arXiv Detail & Related papers (2023-11-29T18:59:59Z) - Dynamic Visual Semantic Sub-Embeddings and Fast Re-Ranking [0.5242869847419834]
We propose a Dynamic Visual Semantic Sub-Embeddings framework (DVSE) to reduce the information entropy.
To encourage the generated candidate embeddings to capture various semantic variations, we construct a mixed distribution.
We compare the performance with existing set-based method using four image feature encoders and two text feature encoders on three benchmark datasets.
arXiv Detail & Related papers (2023-09-15T04:39:11Z) - A Variational Perspective on Solving Inverse Problems with Diffusion
Models [101.831766524264]
Inverse tasks can be formulated as inferring a posterior distribution over data.
This is however challenging in diffusion models since the nonlinear and iterative nature of the diffusion process renders the posterior intractable.
We propose a variational approach that by design seeks to approximate the true posterior distribution.
arXiv Detail & Related papers (2023-05-07T23:00:47Z) - Cross-Modal Discrete Representation Learning [73.68393416984618]
We present a self-supervised learning framework that learns a representation that captures finer levels of granularity across different modalities.
Our framework relies on a discretized embedding space created via vector quantization that is shared across different modalities.
arXiv Detail & Related papers (2021-06-10T00:23:33Z) - Deep Variational Network Toward Blind Image Restoration [60.45350399661175]
Blind image restoration is a common yet challenging problem in computer vision.
We propose a novel blind image restoration method, aiming to integrate both the advantages of them.
Experiments on two typical blind IR tasks, namely image denoising and super-resolution, demonstrate that the proposed method achieves superior performance over current state-of-the-arts.
arXiv Detail & Related papers (2020-08-25T03:30:53Z)
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.