Engineering optimization problems are often nonlinear, high-dimensional, and constrained, making them challenging for conventional optimization techniques. Although L-SHADE, an adaptive differential evolution (DE) algorithm with success-history based parameter adaptation, has demonstrated competitive performance, it still suffer from limited population diversity and weak local exploitation, leading to an imbalance between exploration and exploitation in complex optimization. To address these limitations, this paper proposes LCO-LSHADE-GSRL, a novel DE variant that enhances both global exploration and local exploitation capabilities. The proposed algorithm integrates three key components: (1) a Logistic Chaos Orthogonal Initialization mechanism that improves initial population diversity and ensures uniform coverage of the search space. (2) a GAN-driven Specular Reflection Learning (SRL) mechanism that effectively escapes from local optima. (3) a design that adapts effectively to constrained optimization scenarios. Comprehensive experiments conducted on the CEC 2019 and 2022 benchmark suites demonstrate that LCO-LSHADE-GSRL exhibits superior convergence performance, solution accuracy, and robustness compared to L-SHADE, LSHADE-cnEpSin, and WOA, GJO, PO, PIMO, and CDO. Furthermore, in three real-world engineering problems–speed reducer, step-cone pulley, and hydrostatic thrust bearing, which reduces system weight and power loss while satisfying all design constraints. These results demonstrate its potential for solving complex engineering optimization tasks with high reliability and efficiency.

Recent years have witnessed significant advancements in text-guided style transfer, primarily attributed to innovations in diffusion models. These models excel in conditional guidance, utilizing text or images to direct the sampling process. Traditional style transfer focuses on low-level visual features, such as brushstroke textures and color distributions, and appears more like applying an artistic filter to an image. Artistic attribute transfer, however, transcends the limitations of traditional style transfer by achieving the transfer of visual concepts from color and brushstrokes to high level aesthetic attributes such as composition, pose, and key semantic elements, resulting in more natural outcomes. Therefore, we propose an innovative text-to-image artistic attribute transfer framework named ArtCrafter. Specifically, we introduce an attention-based style extraction module, meticulously engineered to capture the subtle artistic attribute elements within an image. This module features a multi-layer architecture that leverages the capabilities of perceiver attention mechanisms to integrate fine-grained information. Additionally, we present a novel text-image aligning augmentation component that adeptly balances control over both modalities, enabling the model to efficiently map image and text embeddings into a shared feature space. We achieve this through attention operations that enable smooth information flow between modalities. Lastly, we incorporate an explicit modulation that seamlessly blends multimodal enhanced embeddings with original embeddings through an embedding reframing design, empowering the model to generate diverse outputs. Extensive experiments demonstrate that ArtCrafter yields impressive results in visual stylization, exhibiting exceptional levels of artistic attribute intensity, controllability, and diversity.

Unsupervised semantic segmentation algorithms aim to identify meaningful semantic groups without annotations. Recent approaches leveraging self-supervised transformers as pre-training backbones have successfully obtained high-level dense features that effectively express semantic coherence. However, these methods often overlook local semantic coherence and low-level features such as color and texture. We propose integrating low-level visual cues to complement high-level visual cues derived from self-supervised pre-training branches. Our findings indicate that low-level visual cues provide a more coherent recognition of color-texture aspects, ensuring the continuity of spatial structures within classes. This insight led us to develop IL2Vseg, an unsupervised semantic segmentation method that leverages the complementation of low-level visual cues. The core of IL2Vseg is a spatially-constrained fuzzy clustering algorithm based on color affinities, which preserves the intra-class affinity of spatially-adjacent and similarly-colored pixels in low-level visual cues. Additionally, to effectively couple low-level and high-level visual cues, we introduce a feature similarity loss function to optimize the feature representation of fused visual cues. To further enhance consistent feature learning, we incorporate contrast loss functions based on color invariance and luminosity invariance, which improve the learning of features from different semantic categories. Extensive experiments on multiple datasets, including COCO-Stuff-27, Cityscapes, Potsdam, and MaSTr1325, demonstrate that IL2Vseg achieves state-of-the-art results.

Video Object Segmentation (VOS) in the scenario of Low Frame Rate Video (LFR-V) is a promising solution for deploying VOS methods on edge devices with limited computation, storage and transmission bandwidth. However, LFR-V leads robustness challenges that cause existing VOS methods to suffer significant performance degradation. Despite their noisy nature on LFR-V, both motion features and appearance features demonstrate remarkable utility in capturing the dynamic information. How to utilize motion and appearance features more effectively is the key to address the challenges of LFR-V. To this end, we propose a Frame Rate Insensitive Video Object Segmentation (FriVOS) model. In detail, we introduce the Interactive Appearance and Motion (IAM) module to explicitly establish the interactions between appearance and motion. Designs of training are proposed for mitigating the gap of training and inference, further capturing the information on LFR-V. Empirically, sufficient experiments show the effectiveness of our methods in training and inference on LFR-V. The proposed methods surpass other outstanding methods across multiple benchmarks 6.2% at most (e.g., DeAOT on DAVIS-2017). More additional experimental analyses validate the robustness of our methods on LFR-V.

The success of diffusion models in text-to-image (T2I) generation has made it urgent to remove unwanted concepts, such as copyrighted, offensive, and unsafe ones, from pre-trained models in an accurate, timely, and cost-effective manner. However, limited by the inherent optimization perspective, existing methods have two major problems. Firstly, they overlook maintaining the global visual style during the erasure process, leading to significant style shifts. Secondly, excessive concept erasure causes relevant content to disappear or generates substitutes unrelated to the original object's attributes. Compared to other methods, our proposed ICE has unique advantages, as it can generate diverse visual features and achieve a balance between concept erasure and maintaining the semantic content of the target object. This is mainly achieved through our well-designed non-erasable features protector (NEFP) and augmented invariant constraints (AIC). Specifically, we enhance the protection of feature information by embedding an augmented orthogonal anchor concept matrix. Meanwhile, under controlled constraints, we introduce invariants into the embedding space to retain key semantics. This work specifically emphasizes the importance of focusing on feature expression and semantic protection in the concept erasure task for fully unleashing the performance of T2I models.