Distribution Backtracking Builds A Faster Convergence Trajectory for One-step Diffusion Distillation

Accelerating the sampling speed of diffusion models remains a significant challenge. Recent score distillation methods distill a heavy teacher model into an one-step student generator, which is optimized by calculating the difference between the two score functions on the samples generated by the student model. However, there is a score mismatch issue in the early stage of the distillation process, because existing methods mainly focus on using the endpoint of pre-trained diffusion models as teacher models, overlooking the importance of the convergence trajectory between the student generator and the teacher model. To address this issue, we extend the score distillation process by introducing the entire convergence trajectory of teacher models and propose Distribution Backtracking Distillation (DisBack) for distilling student generators. DisBask is composed of two stages: Degradation Recording and Distribution Backtracking. Degradation Recording is designed to obtain the convergence trajectory of teacher models, which records the degradation path from the trained teacher model to the untrained initial student generator. The degradation path implicitly represents the intermediate distributions of teacher models. Then Distribution Backtracking trains a student generator to backtrack the intermediate distributions for approximating the convergence trajectory of teacher models. Extensive experiments show that DisBack achieves faster and better convergence than the existing distillation method and accomplishes comparable generation performance. Notably, DisBack is easy to implement and can be generalized to existing distillation methods to boost performance. Our code is publicly available on https://github.com/SYZhang0805/DisBack.

SAM2-Adapter: Evaluating & Adapting Segment Anything 2 in Downstream Tasks: Camouflage, Shadow, Medical Image Segmentation, and More

The advent of large models, also known as foundation models, has significantly transformed the AI research landscape, with models like Segment Anything (SAM) achieving notable success in diverse image segmentation scenarios. Despite its advancements, SAM encountered limitations in handling some complex low-level segmentation tasks like camouflaged object and medical imaging. In response, in 2023, we introduced SAM-Adapter, which demonstrated improved performance on these challenging tasks. Now, with the release of Segment Anything 2 (SAM2), a successor with enhanced architecture and a larger training corpus, we reassess these challenges. This paper introduces SAM2-Adapter, the first adapter designed to overcome the persistent limitations observed in SAM2 and achieve new state-of-the-art (SOTA) results in specific downstream tasks including medical image segmentation, camouflaged (concealed) object detection, and shadow detection. SAM2-Adapter builds on the SAM-Adapter's strengths, offering enhanced generalizability and composability for diverse applications. We present extensive experimental results demonstrating SAM2-Adapter's effectiveness. We show the potential and encourage the research community to leverage the SAM2 model with our SAM2-Adapter for achieving superior segmentation outcomes. Code, pre-trained models, and data processing protocols are available at http://tianrun-chen.github.io/SAM-Adaptor/

Reasoning3D -- Grounding and Reasoning in 3D: Fine-Grained Zero-Shot Open-Vocabulary 3D Reasoning Part Segmentation via Large Vision-Language Models

In this paper, we introduce a new task: Zero-Shot 3D Reasoning Segmentation for parts searching and localization for objects, which is a new paradigm to 3D segmentation that transcends limitations for previous category-specific 3D semantic segmentation, 3D instance segmentation, and open-vocabulary 3D segmentation. We design a simple baseline method, Reasoning3D, with the capability to understand and execute complex commands for (fine-grained) segmenting specific parts for 3D meshes with contextual awareness and reasoned answers for interactive segmentation. Specifically, Reasoning3D leverages an off-the-shelf pre-trained 2D segmentation network, powered by Large Language Models (LLMs), to interpret user input queries in a zero-shot manner. Previous research have shown that extensive pre-training endows foundation models with prior world knowledge, enabling them to comprehend complex commands, a capability we can harness to "segment anything" in 3D with limited 3D datasets (source efficient). Experimentation reveals that our approach is generalizable and can effectively localize and highlight parts of 3D objects (in 3D mesh) based on implicit textual queries, including these articulated 3d objects and real-world scanned data. Our method can also generate natural language explanations corresponding to these 3D models and the decomposition. Moreover, our training-free approach allows rapid deployment and serves as a viable universal baseline for future research of part-level 3d (semantic) object understanding in various fields including robotics, object manipulation, part assembly, autonomous driving applications, augment reality and virtual reality (AR/VR), and medical applications. The code, the model weight, the deployment guide, and the evaluation protocol are: http://tianrun-chen.github.io/Reason3D/

Efficient and Scalable Chinese Vector Font Generation via Component Composition

Chinese vector font generation is challenging due to the complex structure and huge amount of Chinese characters. Recent advances remain limited to generating a small set of characters with simple structure. In this work, we first observe that most Chinese characters can be disassembled into frequently-reused components. Therefore, we introduce the first efficient and scalable Chinese vector font generation approach via component composition, allowing generating numerous vector characters from a small set of components. To achieve this, we collect a large-scale dataset that contains over \textit{90K} Chinese characters with their components and layout information. Upon the dataset, we propose a simple yet effective framework based on spatial transformer networks (STN) and multiple losses tailored to font characteristics to learn the affine transformation of the components, which can be directly applied to the B\'ezier curves, resulting in Chinese characters in vector format. Our qualitative and quantitative experiments have demonstrated that our method significantly surpasses the state-of-the-art vector font generation methods in generating large-scale complex Chinese characters in both font generation and zero-shot font extension.

ChatScratch: An AI-Augmented System Toward Autonomous Visual Programming Learning for Children Aged 6-12

As Computational Thinking (CT) continues to permeate younger age groups in K-12 education, established CT platforms such as Scratch face challenges in catering to these younger learners, particularly those in the elementary school (ages 6-12). Through formative investigation with Scratch experts, we uncover three key obstacles to children's autonomous Scratch learning: artist's block in project planning, bounded creativity in asset creation, and inadequate coding guidance during implementation. To address these barriers, we introduce ChatScratch, an AI-augmented system to facilitate autonomous programming learning for young children. ChatScratch employs structured interactive storyboards and visual cues to overcome artist's block, integrates digital drawing and advanced image generation technologies to elevate creativity, and leverages Scratch-specialized Large Language Models (LLMs) for professional coding guidance. Our study shows that, compared to Scratch, ChatScratch efficiently fosters autonomous programming learning, and contributes to the creation of high-quality, personally meaningful Scratch projects for children.

Reducing Spatial Fitting Error in Distillation of Denoising Diffusion Models

Denoising Diffusion models have exhibited remarkable capabilities in image generation. However, generating high-quality samples requires a large number of iterations. Knowledge distillation for diffusion models is an effective method to address this limitation with a shortened sampling process but causes degraded generative quality. Based on our analysis with bias-variance decomposition and experimental observations, we attribute the degradation to the spatial fitting error occurring in the training of both the teacher and student model. Accordingly, we propose $\textbf{S}$patial $\textbf{F}$itting-$\textbf{E}$rror $\textbf{R}$eduction $\textbf{D}$istillation model ($\textbf{SFERD}$). SFERD utilizes attention guidance from the teacher model and a designed semantic gradient predictor to reduce the student's fitting error. Empirically, our proposed model facilitates high-quality sample generation in a few function evaluations. We achieve an FID of 5.31 on CIFAR-10 and 9.39 on ImageNet 64$\times$64 with only one step, outperforming existing diffusion methods. Our study provides a new perspective on diffusion distillation by highlighting the intrinsic denoising ability of models.

Deep3DSketch+: Rapid 3D Modeling from Single Free-hand Sketches

The rapid development of AR/VR brings tremendous demands for 3D content. While the widely-used Computer-Aided Design (CAD) method requires a time-consuming and labor-intensive modeling process, sketch-based 3D modeling offers a potential solution as a natural form of computer-human interaction. However, the sparsity and ambiguity of sketches make it challenging to generate high-fidelity content reflecting creators' ideas. Precise drawing from multiple views or strategic step-by-step drawings is often required to tackle the challenge but is not friendly to novice users. In this work, we introduce a novel end-to-end approach, Deep3DSketch+, which performs 3D modeling using only a single free-hand sketch without inputting multiple sketches or view information. Specifically, we introduce a lightweight generation network for efficient inference in real-time and a structural-aware adversarial training approach with a Stroke Enhancement Module (SEM) to capture the structural information to facilitate learning of the realistic and fine-detailed shape structures for high-fidelity performance. Extensive experiments demonstrated the effectiveness of our approach with the state-of-the-art (SOTA) performance on both synthetic and real datasets.

MelodyGLM: Multi-task Pre-training for Symbolic Melody Generation

Pre-trained language models have achieved impressive results in various music understanding and generation tasks. However, existing pre-training methods for symbolic melody generation struggle to capture multi-scale, multi-dimensional structural information in note sequences, due to the domain knowledge discrepancy between text and music. Moreover, the lack of available large-scale symbolic melody datasets limits the pre-training improvement. In this paper, we propose MelodyGLM, a multi-task pre-training framework for generating melodies with long-term structure. We design the melodic n-gram and long span sampling strategies to create local and global blank infilling tasks for modeling the local and global structures in melodies. Specifically, we incorporate pitch n-grams, rhythm n-grams, and their combined n-grams into the melodic n-gram blank infilling tasks for modeling the multi-dimensional structures in melodies. To this end, we have constructed a large-scale symbolic melody dataset, MelodyNet, containing more than 0.4 million melody pieces. MelodyNet is utilized for large-scale pre-training and domain-specific n-gram lexicon construction. Both subjective and objective evaluations demonstrate that MelodyGLM surpasses the standard and previous pre-training methods. In particular, subjective evaluations show that, on the melody continuation task, MelodyGLM gains average improvements of 0.82, 0.87, 0.78, and 0.94 in consistency, rhythmicity, structure, and overall quality, respectively. Notably, MelodyGLM nearly matches the quality of human-composed melodies on the melody inpainting task.

EGFE: End-to-end Grouping of Fragmented Elements in UI Designs with Multimodal Learning

When translating UI design prototypes to code in industry, automatically generating code from design prototypes can expedite the development of applications and GUI iterations. However, in design prototypes without strict design specifications, UI components may be composed of fragmented elements. Grouping these fragmented elements can greatly improve the readability and maintainability of the generated code. Current methods employ a two-stage strategy that introduces hand-crafted rules to group fragmented elements. Unfortunately, the performance of these methods is not satisfying due to visually overlapped and tiny UI elements. In this study, we propose EGFE, a novel method for automatically End-to-end Grouping Fragmented Elements via UI sequence prediction. To facilitate the UI understanding, we innovatively construct a Transformer encoder to model the relationship between the UI elements with multi-modal representation learning. The evaluation on a dataset of 4606 UI prototypes collected from professional UI designers shows that our method outperforms the state-of-the-art baselines in the precision (by 29.75\%), recall (by 31.07\%), and F1-score (by 30.39\%) at edit distance threshold of 4. In addition, we conduct an empirical study to assess the improvement of the generated front-end code. The results demonstrate the effectiveness of our method on a real software engineering application. Our end-to-end fragmented elements grouping method creates opportunities for improving UI-related software engineering tasks.

Efficient Emotional Adaptation for Audio-Driven Talking-Head Generation

Audio-driven talking-head synthesis is a popular research topic for virtual human-related applications. However, the inflexibility and inefficiency of existing methods, which necessitate expensive end-to-end training to transfer emotions from guidance videos to talking-head predictions, are significant limitations. In this work, we propose the Emotional Adaptation for Audio-driven Talking-head (EAT) method, which transforms emotion-agnostic talking-head models into emotion-controllable ones in a cost-effective and efficient manner through parameter-efficient adaptations. Our approach utilizes a pretrained emotion-agnostic talking-head transformer and introduces three lightweight adaptations (the Deep Emotional Prompts, Emotional Deformation Network, and Emotional Adaptation Module) from different perspectives to enable precise and realistic emotion controls. Our experiments demonstrate that our approach achieves state-of-the-art performance on widely-used benchmarks, including LRW and MEAD. Additionally, our parameter-efficient adaptations exhibit remarkable generalization ability, even in scenarios where emotional training videos are scarce or nonexistent. Project website: https://yuangan.github.io/eat/