Jian Jin

Aaron Grattafiori, Abhimanyu Dubey, Abhinav Jauhri et al. · allen-ai, berkeley

Modern artificial intelligence (AI) systems are powered by foundation models. This paper presents a new set of foundation models, called Llama 3. It is a herd of language models that natively support multilinguality, coding, reasoning, and tool usage. Our largest model is a dense Transformer with 405B parameters and a context window of up to 128K tokens. This paper presents an extensive empirical evaluation of Llama 3. We find that Llama 3 delivers comparable quality to leading language models such as GPT-4 on a plethora of tasks. We publicly release Llama 3, including pre-trained and post-trained versions of the 405B parameter language model and our Llama Guard 3 model for input and output safety. The paper also presents the results of experiments in which we integrate image, video, and speech capabilities into Llama 3 via a compositional approach. We observe this approach performs competitively with the state-of-the-art on image, video, and speech recognition tasks. The resulting models are not yet being broadly released as they are still under development.

Jian Jin, Yuan Xue, Xingxing Zhang et al.

Significant improvement has been made on just noticeable difference (JND) modelling due to the development of deep neural networks, especially for the recently developed unsupervised-JND generation models. However, they have a major drawback that the generated JND is assessed in the real-world signal domain instead of in the perceptual domain in the human brain. There is an obvious difference when JND is assessed in such two domains since the visual signal in the real world is encoded before it is delivered into the brain with the human visual system (HVS). Hence, we propose an HVS-inspired signal degradation network for JND estimation. To achieve this, we carefully analyze the HVS perceptual process in JND subjective viewing to obtain relevant insights, and then design an HVS-inspired signal degradation (HVS-SD) network to represent the signal degradation in the HVS. On the one hand, the well learnt HVS-SD enables us to assess the JND in the perceptual domain. On the other hand, it provides more accurate prior information for better guiding JND generation. Additionally, considering the requirement that reasonable JND should not lead to visual attention shifting, a visual attention loss is proposed to control JND generation. Experimental results demonstrate that the proposed method achieves the SOTA performance for accurately estimating the redundancy of the HVS. Source code will be available at https://github.com/jianjin008/HVS-SD-JND.

Wen Sun, Jian Jin, Weisi Lin

Deep learning models are found to be vulnerable to adversarial examples, as wrong predictions can be caused by small perturbation in input for deep learning models. Most of the existing works of adversarial image generation try to achieve attacks for most models, while few of them make efforts on guaranteeing the perceptual quality of the adversarial examples. High quality adversarial examples matter for many applications, especially for the privacy preserving. In this work, we develop a framework based on the Minimum Noticeable Difference (MND) concept to generate adversarial privacy preserving images that have minimum perceptual difference from the clean ones but are able to attack deep learning models. To achieve this, an adversarial loss is firstly proposed to make the deep learning models attacked by the adversarial images successfully. Then, a perceptual quality-preserving loss is developed by taking the magnitude of perturbation and perturbation-caused structural and gradient changes into account, which aims to preserve high perceptual quality for adversarial image generation. To the best of our knowledge, this is the first work on exploring quality-preserving adversarial image generation based on the MND concept for privacy preserving. To evaluate its performance in terms of perceptual quality, the deep models on image classification and face recognition are tested with the proposed method and several anchor methods in this work. Extensive experimental results demonstrate that the proposed MND framework is capable of generating adversarial images with remarkably improved performance metrics (e.g., PSNR, SSIM, and MOS) than that generated with the anchor methods.

Jian Jin, Dong Yu, Weisi Lin et al.

Just Noticeable Difference (JND) has many applications in multimedia signal processing, especially for visual data processing up to date. It's generally defined as the minimum visual content changes that the human can perspective, which has been studied for decades. However, most of the existing methods only focus on the luminance component of JND modelling and simply regard chrominance components as scaled versions of luminance. In this paper, we propose a JND model to generate the JND by taking the characteristics of full RGB channels into account, termed as the RGB-JND. To this end, an RGB-JND-NET is proposed, where the visual content in full RGB channels is used to extract features for JND generation. To supervise the JND generation, an adaptive image quality assessment combination (AIC) is developed. Besides, the RDB-JND-NET also takes the visual attention into account by automatically mining the underlying relationship between visual attention and the JND, which is further used to constrain the JND spatial distribution. To the best of our knowledge, this is the first work on careful investigation of JND modelling for full-color space. Experimental results demonstrate that the RGB-JND-NET model outperforms the relevant state-of-the-art JND models. Besides, the JND of the red and blue channels are larger than that of the green one according to the experimental results of the proposed model, which demonstrates that more changes can be tolerated in the red and blue channels, in line with the well-known fact that the human visual system is more sensitive to the green channel in comparison with the red and blue ones.

Feng Ding, Jian Jin, Lili Meng et al.

Recently, learned image compression schemes have achieved remarkable improvements in image fidelity (e.g., PSNR and MS-SSIM) compared to conventional hybrid image coding ones due to their high-efficiency non-linear transform, end-to-end optimization frameworks, etc. However, few of them take the Just Noticeable Difference (JND) characteristic of the Human Visual System (HVS) into account and optimize learned image compression towards perceptual quality. To address this issue, a JND-based perceptual quality loss is proposed. Considering that the amounts of distortion in the compressed image at different training epochs under different Quantization Parameters (QPs) are different, we develop a distortion-aware adjustor. After combining them together, we can better assign the distortion in the compressed image with the guidance of JND to preserve the high perceptual quality. All these designs enable the proposed method to be flexibly applied to various learned image compression schemes with high scalability and plug-and-play advantages. Experimental results on the Kodak dataset demonstrate that the proposed method has led to better perceptual quality than the baseline model under the same bit rate.

Yaxuan Liu, Jian Jin, Yuan Xue et al.

Recently, with the development of deep learning, a number of Just Noticeable Difference (JND) datasets have been built for JND modeling. However, all the existing JND datasets only label the JND points based on the level of compression distortion. Hence, JND models learned from such datasets can only be used for image/video compression. As known, JND is a major characteristic of the human visual system (HVS), which reflects the maximum visual distortion that the HVS can tolerate. Hence, a generalized JND modeling should take more kinds of distortion types into account. To benefit JND modeling, this work establishes a generalized JND dataset with a coarse-to-fine JND selection, which contains 106 source images and 1,642 JND maps, covering 25 distortion types. To this end, we proposed a coarse JND candidate selection scheme to select the distorted images from the existing Image Quality Assessment (IQA) datasets as JND candidates instead of generating JND maps ourselves. Then, a fine JND selection is carried out on the JND candidates with a crowdsourced subjective assessment.

Renyu Yang, Jian Jin, Lili Meng et al.

Audio-visual quality assessment (AVQA) research has been stalled by limitations of existing datasets: they are typically small in scale, with insufficient diversity in content and quality, and annotated only with overall scores. These shortcomings provide limited support for model development and multimodal perception research. We propose a practical approach for AVQA dataset construction. First, we design a crowdsourced subjective experiment framework for AVQA, breaks the constraints of in-lab settings and achieves reliable annotation across varied environments. Second, a systematic data preparation strategy is further employed to ensure broad coverage of both quality levels and semantic scenarios. Third, we extend the dataset with additional annotations, enabling research on multimodal perception mechanisms and their relation to content. Finally, we validate this approach through YT-NTU-AVQ, the largest and most diverse AVQA dataset to date, consisting of 1,620 user-generated audio and video (A/V) sequences. The dataset and platform code are available at https://github.com/renyu12/YT-NTU-AVQ

Guohua Zhang, Jian Jin, Meiqin Liu et al.

No-Reference Point Cloud Quality Assessment (NR-PCQA) still struggles with generalization, primarily due to the scarcity of annotated point cloud datasets. Since the Human Visual System (HVS) drives perceptual quality assessment independently of media types, prior knowledge on quality learned from images can be repurposed for point clouds. This insight motivates adopting Unsupervised Domain Adaptation (UDA) to transfer quality-relevant priors from labeled images to unlabeled point clouds. However, existing UDA-based PCQA methods often overlook key characteristics of perceptual quality, such as sensitivity to quality ranking and quality-aware feature alignment, thereby limiting their effectiveness. To address these issues, we propose a novel Quality-aware Domain adaptation framework for PCQA, termed QD-PCQA. The framework comprises two main components: i) a Rank-weighted Conditional Alignment (RCA) strategy that aligns features under consistent quality levels and adaptively emphasizes misranked samples to reinforce perceptual quality ranking awareness; and ii) a Quality-guided Feature Augmentation (QFA) strategy, which includes quality-guided style mixup, multi-layer extension, and dual-domain augmentation modules to augment perceptual feature alignment. Extensive cross-domain experiments demonstrate that QD-PCQA significantly improves generalization in NR-PCQA tasks. The code is available at https://github.com/huhu-code/QD-PCQA.

Jiaju Lin, Qin Chen, Jie Zhou et al.

Implicit event argument extraction (EAE) aims to identify arguments that could scatter over the document. Most previous work focuses on learning the direct relations between arguments and the given trigger, while the implicit relations with long-range dependency are not well studied. Moreover, recent neural network based approaches rely on a large amount of labeled data for training, which is unavailable due to the high labelling cost. In this paper, we propose a Curriculum learning based Prompt tuning (CUP) approach, which resolves implicit EAE by four learning stages. The stages are defined according to the relations with the trigger node in a semantic graph, which well captures the long-range dependency between arguments and the trigger. In addition, we integrate a prompt-based encoder-decoder model to elicit related knowledge from pre-trained language models (PLMs) in each stage, where the prompt templates are adapted with the learning progress to enhance the reasoning for arguments. Experimental results on two well-known benchmark datasets show the great advantages of our proposed approach. In particular, we outperform the state-of-the-art models in both fully-supervised and low-data scenarios.

Tongtong Yuan, Xuange Zhang, Kun Liu et al.

Surveillance videos are an essential component of daily life with various critical applications, particularly in public security. However, current surveillance video tasks mainly focus on classifying and localizing anomalous events. Existing methods are limited to detecting and classifying the predefined events with unsatisfactory semantic understanding, although they have obtained considerable performance. To address this issue, we propose a new research direction of surveillance video-and-language understanding, and construct the first multimodal surveillance video dataset. We manually annotate the real-world surveillance dataset UCF-Crime with fine-grained event content and timing. Our newly annotated dataset, UCA (UCF-Crime Annotation), contains 23,542 sentences, with an average length of 20 words, and its annotated videos are as long as 110.7 hours. Furthermore, we benchmark SOTA models for four multimodal tasks on this newly created dataset, which serve as new baselines for surveillance video-and-language understanding. Through our experiments, we find that mainstream models used in previously publicly available datasets perform poorly on surveillance video, which demonstrates the new challenges in surveillance video-and-language understanding. To validate the effectiveness of our UCA, we conducted experiments on multimodal anomaly detection. The results demonstrate that our multimodal surveillance learning can improve the performance of conventional anomaly detection tasks. All the experiments highlight the necessity of constructing this dataset to advance surveillance AI. The link to our dataset is provided at: https://xuange923.github.io/Surveillance-Video-Understanding.

Fanxin Xia, Jian Jin, Lili Meng et al.

GAN-based image compression schemes have shown remarkable progress lately due to their high perceptual quality at low bit rates. However, there are two main issues, including 1) the reconstructed image perceptual degeneration in color, texture, and structure as well as 2) the inaccurate entropy model. In this paper, we present a novel GAN-based image compression approach with improved rate-distortion optimization (RDO) process. To achieve this, we utilize the DISTS and MS-SSIM metrics to measure perceptual degeneration in color, texture, and structure. Besides, we absorb the discretized gaussian-laplacian-logistic mixture model (GLLMM) for entropy modeling to improve the accuracy in estimating the probability distributions of the latent representation. During the evaluation process, instead of evaluating the perceptual quality of the reconstructed image via IQA metrics, we directly conduct the Mean Opinion Score (MOS) experiment among different codecs, which fully reflects the actual perceptual results of humans. Experimental results demonstrate that the proposed method outperforms the existing GAN-based methods and the state-of-the-art hybrid codec (i.e., VVC).

Guohua Zhang, Jian Jin, Meiqin Liu et al.

With the rapid advancement of Multi-modal Large Language Models (MLLMs), MLLM-based Image Quality Assessment (IQA) methods have shown promising generalization. However, directly extending these MLLM-based IQA methods to PCQA remains challenging. On the one hand, existing PCQA datasets are limited in scale, which hinders stable and effective instruction tuning of MLLMs. On the other hand, due to large-scale image-text pretraining, MLLMs tend to rely on texture-dominant reasoning and are insufficiently sensitive to geometric structural degradations that are critical for PCQA. To address these gaps, we propose a novel MLLM-based no-reference PCQA framework, termed GT-PCQA, which is built upon two key strategies. First, to enable stable and effective instruction tuning under scarce PCQA supervision, a 2D-3D joint training strategy is proposed. This strategy formulates PCQA as a relative quality comparison problem to unify large-scale IQA datasets with limited PCQA datasets. It incorporates a parameter-efficient Low-Rank Adaptation (LoRA) scheme to support instruction tuning. Second, a geometry-texture decoupling strategy is presented, which integrates a dual-prompt mechanism with an alternating optimization scheme to mitigate the inherent texture-dominant bias of pre-trained MLLMs, while enhancing sensitivity to geometric structural degradations. Extensive experiments demonstrate that GT-PCQA achieves competitive performance and exhibits strong generalization.

Tianang Chen, Jian Jin, Shilv Cai et al.

Gaussian Splatting (GS) has recently emerged as a promising technique for 3D object reconstruction, delivering high-quality rendering results with significantly improved reconstruction speed. As variants continue to appear, assessing the perceptual quality of 3D objects reconstructed with different GS-based methods remains an open challenge. To address this issue, we first propose a unified multi-distance subjective quality assessment method that closely mimics human viewing behavior for objects reconstructed with GS-based methods in actual applications, thereby better collecting perceptual experiences. Based on it, we also construct a novel GS quality assessment dataset named MUGSQA, which is constructed considering multiple uncertainties of the input data. These uncertainties include the quantity and resolution of input views, the view distance, and the accuracy of the initial point cloud. Moreover, we construct two benchmarks: one to evaluate the robustness of various GS-based reconstruction methods under multiple uncertainties, and the other to evaluate the performance of existing quality assessment metrics. Our dataset and benchmark code will be released soon.

Jian Jin, Zhenbo Yu, Yang Shen et al.

Customized text-to-image generation renders user-specified concepts into novel contexts based on textual prompts. Scaling the number of concepts in customized generation meets a broader demand for user creation, whereas existing methods face challenges with generation quality and computational efficiency. In this paper, we propose LaTexBlend, a novel framework for effectively and efficiently scaling multi-concept customized generation. The core idea of LaTexBlend is to represent single concepts and blend multiple concepts within a Latent Textual space, which is positioned after the text encoder and a linear projection. LaTexBlend customizes each concept individually, storing them in a concept bank with a compact representation of latent textual features that captures sufficient concept information to ensure high fidelity. At inference, concepts from the bank can be freely and seamlessly combined in the latent textual space, offering two key merits for multi-concept generation: 1) excellent scalability, and 2) significant reduction of denoising deviation, preserving coherent layouts. Extensive experiments demonstrate that LaTexBlend can flexibly integrate multiple customized concepts with harmonious structures and high subject fidelity, substantially outperforming baselines in both generation quality and computational efficiency. Our code will be publicly available.

Yuan Yao, Jin Song, Jian Jin

As valuable digital assets, deep neural networks necessitate robust ownership protection, positioning neural network watermarking (NNW) as a promising solution. Among various NNW approaches, weight-based methods are favored for their simplicity and practicality; however, they remain vulnerable to forging and overwriting attacks. To address those challenges, we propose NeuralMark, a robust method built around a hashed watermark filter. Specifically, we utilize a hash function to generate an irreversible binary watermark from a secret key, which is then used as a filter to select the model parameters for embedding. This design cleverly intertwines the embedding parameters with the hashed watermark, providing a robust defense against both forging and overwriting attacks. An average pooling is also incorporated to resist fine-tuning and pruning attacks. Furthermore, it can be seamlessly integrated into various neural network architectures, ensuring broad applicability. Theoretically, we analyze its security boundary. Empirically, we verify its effectiveness and robustness across 13 distinct Convolutional and Transformer architectures, covering five image classification tasks and one text generation task. The source codes are available at https://github.com/AIResearch-Group/NeuralMark.

Yulong Zhang, Yuan Yao, Shuhao Chen et al.

Empirical Risk Minimization (ERM) is fragile in scenarios with insufficient labeled samples. A vanilla extension of ERM to unlabeled samples is Entropy Minimization (EntMin), which employs the soft-labels of unlabeled samples to guide their learning. However, EntMin emphasizes prediction discriminability while neglecting prediction diversity. To alleviate this issue, in this paper, we rethink the guidance information to utilize unlabeled samples. By analyzing the learning objective of ERM, we find that the guidance information for labeled samples in a specific category is the corresponding label encoding. Inspired by this finding, we propose a Label-Encoding Risk Minimization (LERM). It first estimates the label encodings through prediction means of unlabeled samples and then aligns them with their corresponding ground-truth label encodings. As a result, the LERM ensures both prediction discriminability and diversity, and it can be integrated into existing methods as a plugin. Theoretically, we analyze the relationships between LERM and ERM as well as EntMin. Empirically, we verify the superiority of the LERM under several label insufficient scenarios. The codes are available at https://github.com/zhangyl660/LERM.

Yuan Yao, Xiaopu Zhang, Yu Zhang et al.

Semi-supervised heterogeneous domain adaptation (SHDA) addresses learning across domains with distinct feature representations and distributions, where source samples are labeled while most target samples are unlabeled, with only a small fraction labeled. Moreover, there is no one-to-one correspondence between source and target samples. Although various SHDA methods have been developed to tackle this problem, the nature of the knowledge transferred across heterogeneous domains remains unclear. This paper delves into this question from an empirical perspective. We conduct extensive experiments on about 330 SHDA tasks, employing two supervised learning methods and seven representative SHDA methods. Surprisingly, our observations indicate that both the category and feature information of source samples do not significantly impact the performance of the target domain. Additionally, noise drawn from simple distributions, when used as source samples, may contain transferable knowledge. Based on this insight, we perform a series of experiments to uncover the underlying principles of transferable knowledge in SHDA. Specifically, we design a unified Knowledge Transfer Framework (KTF) for SHDA. Based on the KTF, we find that the transferable knowledge in SHDA primarily stems from the transferability and discriminability of the source domain. Consequently, ensuring those properties in source samples, regardless of their origin (e.g., image, text, noise), can enhance the effectiveness of knowledge transfer in SHDA tasks. The codes and datasets are available at https://github.com/yyyaoyuan/SHDA.

Jian Jin, Xingxing Zhang, Lili Meng et al.

Recently, various view synthesis distortion estimation models have been studied to better serve for 3-D video coding. However, they can hardly model the relationship quantitatively among different levels of depth changes, texture degeneration, and the view synthesis distortion (VSD), which is crucial for rate-distortion optimization and rate allocation. In this paper, an auto-weighted layer representation based view synthesis distortion estimation model is developed. Firstly, the sub-VSD (S-VSD) is defined according to the level of depth changes and their associated texture degeneration. After that, a set of theoretical derivations demonstrate that the VSD can be approximately decomposed into the S-VSDs multiplied by their associated weights. To obtain the S-VSDs, a layer-based representation of S-VSD is developed, where all the pixels with the same level of depth changes are represented with a layer to enable efficient S-VSD calculation at the layer level. Meanwhile, a nonlinear mapping function is learnt to accurately represent the relationship between the VSD and S-VSDs, automatically providing weights for S-VSDs during the VSD estimation. To learn such function, a dataset of VSD and its associated S-VSDs are built. Experimental results show that the VSD can be accurately estimated with the weights learnt by the nonlinear mapping function once its associated S-VSDs are available. The proposed method outperforms the relevant state-of-the-art methods in both accuracy and efficiency. The dataset and source code of the proposed method will be available at https://github.com/jianjin008/.

Qi Tang, Yao Zhao, Meiqin Liu et al.

As a critical clue of video super-resolution (VSR), inter-frame alignment significantly impacts overall performance. However, accurate pixel-level alignment is a challenging task due to the intricate motion interweaving in the video. In response to this issue, we introduce a novel paradigm for VSR named Semantic Lens, predicated on semantic priors drawn from degraded videos. Specifically, video is modeled as instances, events, and scenes via a Semantic Extractor. Those semantics assist the Pixel Enhancer in understanding the recovered contents and generating more realistic visual results. The distilled global semantics embody the scene information of each frame, while the instance-specific semantics assemble the spatial-temporal contexts related to each instance. Furthermore, we devise a Semantics-Powered Attention Cross-Embedding (SPACE) block to bridge the pixel-level features with semantic knowledge, composed of a Global Perspective Shifter (GPS) and an Instance-Specific Semantic Embedding Encoder (ISEE). Concretely, the GPS module generates pairs of affine transformation parameters for pixel-level feature modulation conditioned on global semantics. After that, the ISEE module harnesses the attention mechanism to align the adjacent frames in the instance-centric semantic space. In addition, we incorporate a simple yet effective pre-alignment module to alleviate the difficulty of model training. Extensive experiments demonstrate the superiority of our model over existing state-of-the-art VSR methods.

Zhuangzi Li, Jian Jin, Shilv Cai et al.

Immersive Computer Graphics (CGs) rendering has become ubiquitous in modern daily life. However, comprehensively evaluating CG quality remains challenging for two reasons: First, existing CG datasets lack systematic descriptions of rendering quality; and second existing CG quality assessment methods cannot provide reasonable text-based explanations. To address these issues, we first identify six key perceptual dimensions of CG quality from the user perspective and construct a dataset of 3500 CG images with corresponding quality descriptions. Each description covers CG style, content, and perceived quality along the selected dimensions. Furthermore, we use a subset of the dataset to build several question-answer benchmarks based on the descriptions in order to evaluate the responses of existing Vision Language Models (VLMs). We find that current VLMs are not sufficiently accurate in judging fine-grained CG quality, but that descriptions of visually similar images can significantly improve a VLM's understanding of a given CG image. Motivated by this observation, we adopt retrieval-augmented generation and propose a two-stream retrieval framework that effectively enhances the CG quality assessment capabilities of VLMs. Experiments on several representative VLMs demonstrate that our method substantially improves their performance on CG quality assessment.

Baiyan Zhang, Qin Chen, Jie Zhou et al.

Document-level Event Causality Identification (DECI) aims to identify causal relations between two events in documents. Recent research tends to use pre-trained language models to generate the event causal relations. Whereas, these methods are prone to the errors of sequential generation due to multiple events in a document. Moreover, the potential structures such as event coreference and related causal chain are neglected. In this paper, we propose a multi-task learning framework to enhance event causality identification with rationale and structure-aware causal question answering. Specifically, the DECI task is transformed into multiple-choice question answering, and the causes and effects of the questioned event are generated with large language models. In addition, we generate the rationales to explain why these events have causal relations. Moreover, we construct an event structure graph, which models the multi-hop potential relations for causal reasoning of the current event. Experiments on two benchmark datasets show the great advantages of our proposed approach compared to the state-of-the-art methods. Moreover, we conduct both quantitative and qualitative analyses, which shed light on why each component of our approach can lead to great improvements.

Jian Jin, Fanxin Xia, Feng Ding et al.

Region of Interest (ROI)-based image compression optimizes bit allocation by prioritizing ROI for higher-quality reconstruction. However, as the users (including human clients and downstream machine tasks) become more diverse, ROI-based image compression needs to be customizable to support various preferences. For example, different users may define distinct ROI or require different quality trade-offs between ROI and non-ROI. Existing ROI-based image compression schemes predefine the ROI, making it unchangeable, and lack effective mechanisms to balance reconstruction quality between ROI and non-ROI. This work proposes a paradigm for customizable ROI-based deep image compression. First, we develop a Text-controlled Mask Acquisition (TMA) module, which allows users to easily customize their ROI for compression by just inputting the corresponding semantic \emph{text}. It makes the encoder controlled by text. Second, we design a Customizable Value Assign (CVA) mechanism, which masks the non-ROI with a changeable extent decided by users instead of a constant one to manage the reconstruction quality trade-off between ROI and non-ROI. Finally, we present a Latent Mask Attention (LMA) module, where the latent spatial prior of the mask and the latent Rate-Distortion Optimization (RDO) prior of the image are extracted and fused in the latent space, and further used to optimize the latent representation of the source image. Experimental results demonstrate that our proposed customizable ROI-based deep image compression paradigm effectively addresses the needs of customization for ROI definition and mask acquisition as well as the reconstruction quality trade-off management between the ROI and non-ROI.

Jian Jin, Yang Shen, Zhenyong Fu et al.

Customized generation aims to incorporate a novel concept into a pre-trained text-to-image model, enabling new generations of the concept in novel contexts guided by textual prompts. However, customized generation suffers from an inherent trade-off between concept fidelity and editability, i.e., between precisely modeling the concept and faithfully adhering to the prompts. Previous methods reluctantly seek a compromise and struggle to achieve both high concept fidelity and ideal prompt alignment simultaneously. In this paper, we propose a Divide, Conquer, then Integrate (DCI) framework, which performs a surgical adjustment in the early stage of denoising to liberate the fine-tuned model from the fidelity-editability trade-off at inference. The two conflicting components in the trade-off are decoupled and individually conquered by two collaborative branches, which are then selectively integrated to preserve high concept fidelity while achieving faithful prompt adherence. To obtain a better fine-tuned model, we introduce an Image-specific Context Optimization} (ICO) strategy for model customization. ICO replaces manual prompt templates with learnable image-specific contexts, providing an adaptive and precise fine-tuning direction to promote the overall performance. Extensive experiments demonstrate the effectiveness of our method in reconciling the fidelity-editability trade-off.

Yachun Mi, Xingyang He, Shixin Sun et al.

In the digital age, advanced image editing tools pose a serious threat to the integrity of visual content, making image forgery detection and localization a key research focus. Most existing Image Manipulation Localization (IML) methods rely on discriminative learning and require large, high-quality annotated datasets. However, current datasets lack sufficient scale and diversity, limiting model performance in real-world scenarios. To overcome this, recent studies have explored Constrained IML (CIML), which generates pixel-level annotations through algorithmic supervision. However, existing CIML approaches often depend on complex multi-stage pipelines, making the annotation process inefficient. In this work, we propose a novel generative framework based on diffusion models, named UGD-IML, which for the first time unifies both IML and CIML tasks within a single framework. By learning the underlying data distribution, generative diffusion models inherently reduce the reliance on large-scale labeled datasets, allowing our approach to perform effectively even under limited data conditions. In addition, by leveraging a class embedding mechanism and a parameter-sharing design, our model seamlessly switches between IML and CIML modes without extra components or training overhead. Furthermore, the end-to-end design enables our model to avoid cumbersome steps in the data annotation process. Extensive experimental results on multiple datasets demonstrate that UGD-IML outperforms the SOTA methods by an average of 9.66 and 4.36 in terms of F1 metrics for IML and CIML tasks, respectively. Moreover, the proposed method also excels in uncertainty estimation, visualization and robustness.

Xin-yang Zhao, Jian Jin, Yang-yang Li et al.

The Coarse-to-Fine Few-Shot (C2FS) task is designed to train models using only coarse labels, then leverages a limited number of subclass samples to achieve fine-grained recognition capabilities. This task presents two main challenges: coarse-grained supervised pre-training suppresses the extraction of critical fine-grained features for subcategory discrimination, and models suffer from overfitting due to biased distributions caused by limited fine-grained samples. In this paper, we propose the Twofold Debiasing (TFB) method, which addresses these challenges through detailed feature enhancement and distribution calibration. Specifically, we introduce a multi-layer feature fusion reconstruction module and an intermediate layer feature alignment module to combat the model's tendency to focus on simple predictive features directly related to coarse-grained supervision, while neglecting complex fine-grained level details. Furthermore, we mitigate the biased distributions learned by the fine-grained classifier using readily available coarse-grained sample embeddings enriched with fine-grained information. Extensive experiments conducted on five benchmark datasets demonstrate the efficacy of our approach, achieving state-of-the-art results that surpass competitive methods.

Yang Shen, Xiu-Shen Wei, Yifan Sun et al.

Computer Vision (CV) has yet to fully achieve the zero-shot task generalization observed in Natural Language Processing (NLP), despite following many of the milestones established in NLP, such as large transformer models, extensive pre-training, and the auto-regression paradigm, among others. In this paper, we explore the idea that CV adopts discrete and terminological task definitions (\eg, ``image segmentation''), which may be a key barrier to zero-shot task generalization. Our hypothesis is that without truly understanding previously-seen tasks--due to these terminological definitions--deep models struggle to generalize to novel tasks. To verify this, we introduce Explanatory Instructions, which provide an intuitive way to define CV task objectives through detailed linguistic transformations from input images to outputs. We create a large-scale dataset comprising 12 million ``image input $\to$ explanatory instruction $\to$ output'' triplets, and train an auto-regressive-based vision-language model (AR-based VLM) that takes both images and explanatory instructions as input. By learning to follow these instructions, the AR-based VLM achieves instruction-level zero-shot capabilities for previously-seen tasks and demonstrates strong zero-shot generalization for unseen CV tasks. Code and dataset will be openly available on our GitHub repository.

Sien Chen, Jian Jin, Lili Meng et al.

With the AI of Things (AIoT) development, a huge amount of visual data, e.g., images and videos, are produced in our daily work and life. These visual data are not only used for human viewing or understanding but also for machine analysis or decision-making, e.g., intelligent surveillance, automated vehicles, and many other smart city applications. To this end, a new image codec paradigm for both human and machine uses is proposed in this work. Firstly, the high-level instance segmentation map and the low-level signal features are extracted with neural networks. Then, the instance segmentation map is further represented as a profile with the proposed 16-bit gray-scale representation. After that, both 16-bit gray-scale profile and signal features are encoded with a lossless codec. Meanwhile, an image predictor is designed and trained to achieve the general-quality image reconstruction with the 16-bit gray-scale profile and signal features. Finally, the residual map between the original image and the predicted one is compressed with a lossy codec, used for high-quality image reconstruction. With such designs, on the one hand, we can achieve scalable image compression to meet the requirements of different human consumption; on the other hand, we can directly achieve several machine vision tasks at the decoder side with the decoded 16-bit gray-scale profile, e.g., object classification, detection, and segmentation. Experimental results show that the proposed codec achieves comparable results as most learning-based codecs and outperforms the traditional codecs (e.g., BPG and JPEG2000) in terms of PSNR and MS-SSIM for image reconstruction. At the same time, it outperforms the existing codecs in terms of the mAP for object detection and segmentation.

Jian Jin, Xingxing Zhang, Xin Fu et al.

As an important perceptual characteristic of the Human Visual System (HVS), the Just Noticeable Difference (JND) has been studied for decades with image and video processing (e.g., perceptual visual signal compression). However, there is little exploration on the existence of JND for the Deep Machine Vision (DMV), although the DMV has made great strides in many machine vision tasks. In this paper, we take an initial attempt, and demonstrate that the DMV has the JND, termed as the DMV-JND. We then propose a JND model for the image classification task in the DMV. It has been discovered that the DMV can tolerate distorted images with average PSNR of only 9.56dB (the lower the better), by generating JND via unsupervised learning with the proposed DMV-JND-NET. In particular, a semantic-guided redundancy assessment strategy is designed to restrain the magnitude and spatial distribution of the DMV-JND. Experimental results on image classification demonstrate that we successfully find the JND for deep machine vision. Our DMV-JND facilitates a possible direction for DMV-oriented image and video compression, watermarking, quality assessment, deep neural network security, and so on.

Jian Jin

MDL, Multimodal Deep Learning Library, is a deep learning framework that supports multiple models, and this document explains its philosophy and functionality. MDL runs on Linux, Mac, and Unix platforms. It depends on OpenCV.