Ziheng Jia

Zicheng Zhang, Ziheng Jia, Haoning Wu et al.

With the rising interest in research on Large Multi-modal Models (LMMs) for video understanding, many studies have emphasized general video comprehension capabilities, neglecting the systematic exploration into video quality understanding. To address this oversight, we introduce Q-Bench-Video in this paper, a new benchmark specifically designed to evaluate LMMs' proficiency in discerning video quality. a) To ensure video source diversity, Q-Bench-Video encompasses videos from natural scenes, AI-generated Content (AIGC), and Computer Graphics (CG). b) Building on the traditional multiple-choice questions format with the Yes-or-No and What-How categories, we include Open-ended questions to better evaluate complex scenarios. Additionally, we incorporate the video pair quality comparison question to enhance comprehensiveness. c) Beyond the traditional Technical, Aesthetic, and Temporal distortions, we have expanded our evaluation aspects to include the dimension of AIGC distortions, which addresses the increasing demand for video generation. Finally, we collect a total of 2,378 question-answer pairs and test them on 12 open-source & 5 proprietary LMMs. Our findings indicate that while LMMs have a foundational understanding of video quality, their performance remains incomplete and imprecise, with a notable discrepancy compared to human performance. Through Q-Bench-Video, we seek to catalyze community interest, stimulate further research, and unlock the untapped potential of LMMs to close the gap in video quality understanding.

Xiaohong Liu, Xiongkuo Min, Wei Sun et al. · eth-zurich

This paper reports on the NTIRE 2023 Quality Assessment of Video Enhancement Challenge, which will be held in conjunction with the New Trends in Image Restoration and Enhancement Workshop (NTIRE) at CVPR 2023. This challenge is to address a major challenge in the field of video processing, namely, video quality assessment (VQA) for enhanced videos. The challenge uses the VQA Dataset for Perceptual Video Enhancement (VDPVE), which has a total of 1211 enhanced videos, including 600 videos with color, brightness, and contrast enhancements, 310 videos with deblurring, and 301 deshaked videos. The challenge has a total of 167 registered participants. 61 participating teams submitted their prediction results during the development phase, with a total of 3168 submissions. A total of 176 submissions were submitted by 37 participating teams during the final testing phase. Finally, 19 participating teams submitted their models and fact sheets, and detailed the methods they used. Some methods have achieved better results than baseline methods, and the winning methods have demonstrated superior prediction performance.

Andrey Moskalenko, Alexey Bryncev, Dmitry Vatolin et al.

This paper reviews the Challenge on Video Saliency Prediction at AIM 2024. The goal of the participants was to develop a method for predicting accurate saliency maps for the provided set of video sequences. Saliency maps are widely exploited in various applications, including video compression, quality assessment, visual perception studies, the advertising industry, etc. For this competition, a previously unused large-scale audio-visual mouse saliency (AViMoS) dataset of 1500 videos with more than 70 observers per video was collected using crowdsourced mouse tracking. The dataset collection methodology has been validated using conventional eye-tracking data and has shown high consistency. Over 30 teams registered in the challenge, and there are 7 teams that submitted the results in the final phase. The final phase solutions were tested and ranked by commonly used quality metrics on a private test subset. The results of this evaluation and the descriptions of the solutions are presented in this report. All data, including the private test subset, is made publicly available on the challenge homepage - https://challenges.videoprocessing.ai/challenges/video-saliency-prediction.html.

Maksim Smirnov, Aleksandr Gushchin, Anastasia Antsiferova et al.

Video quality assessment (VQA) is a crucial task in the development of video compression standards, as it directly impacts the viewer experience. This paper presents the results of the Compressed Video Quality Assessment challenge, held in conjunction with the Advances in Image Manipulation (AIM) workshop at ECCV 2024. The challenge aimed to evaluate the performance of VQA methods on a diverse dataset of 459 videos, encoded with 14 codecs of various compression standards (AVC/H.264, HEVC/H.265, AV1, and VVC/H.266) and containing a comprehensive collection of compression artifacts. To measure the methods performance, we employed traditional correlation coefficients between their predictions and subjective scores, which were collected via large-scale crowdsourced pairwise human comparisons. For training purposes, participants were provided with the Compressed Video Quality Assessment Dataset (CVQAD), a previously developed dataset of 1022 videos. Up to 30 participating teams registered for the challenge, while we report the results of 6 teams, which submitted valid final solutions and code for reproducing the results. Moreover, we calculated and present the performance of state-of-the-art VQA methods on the developed dataset, providing a comprehensive benchmark for future research. The dataset, results, and online leaderboard are publicly available at https://challenges.videoprocessing.ai/challenges/compressedvideo-quality-assessment.html.

Jiaying Qian, Ziheng Jia, Qian Zhang et al.

With the continued advancement of text-to-image (T2I) generation, producing high-quality images is becoming increasingly attainable; consequently, user demands are shifting toward images that better satisfy their specific requirements. As reward models play an increasingly important role in assessing whether generated images align with user preference, this trend introduces an important challenge for reward modeling: rather than relying solely on static and general evaluation dimensions, reward models should account for the task-relevant and fine-grained criteria through which users assess whether generated images meet their specific requirements. To address this challenge, we propose DyCoRM, a dynamic, criterion-aware reward model that grounds task-relevant criteria and performs criterion-aware preference comparison. To support this setting, we construct DyCoDataset-20K, which provides dynamic criteria together with criterion-level annotations, and further derive DyCoBench-1K, a benchmark for systematically evaluating reward models under dynamic criteria. We further introduce DyCoPick, which applies criterion-aware reward modeling to selecting T2I images. Our contributions establish the first reward modeling framework for dynamic and fine-grained evaluation and practical application in T2I generation.

Yushuo Zheng, Zicheng Zhang, Huiyu Duan et al.

Geoscience intelligence is expected to understand, reason about, and predict earth system changes to support human decision-making in critical domains such as disaster response, climate adaptation and environmental protection. Although current research has shown promising progress on specific geoscience tasks, such as remote sensing interpretation, geographic question-answering, existing benchmarks remain largely task-specific which failing to capture the open-ended real world geoscience problems. As a result, it remains unclear how far current AI systems are from achieving genuine geoscience intelligence. To address this gap, we present \textbf{GeoR-Bench}, a \underline{Bench}mark for evaluating \underline{Geo}science visual \underline{R}easoning through reasoning informed visual editing tasks. GeoR-Bench contains 440 curated samples spanning 6 geoscience categories and 24 task types, covering earth observation imagery and structured scientific representations such as maps and diagrams. We evaluate outputs along three dimensions, including reasoning, consistency, and quality. Benchmark results of 21 closed- and open-source multimodal models reveal that geoscience reasoning remains a critical bottleneck. The highest-performing model achieves 42.7\% overall strict accuracy, while the best open-source models only get 10.3\%. Notably, the visual consistency and image quality of the outputs frequently surpass their scientific accuracy. Ultimately, these findings indicate that current models generate superficially plausible results but fail to capture underlying earth science processes.

Chunyi Li, Yuan Tian, Xiaoyue Ling et al.

Image Quality Assessment (IQA) based on human subjective preferences has undergone extensive research in the past decades. However, with the development of communication protocols, the visual data consumption volume of machines has gradually surpassed that of humans. For machines, the preference depends on downstream tasks such as segmentation and detection, rather than visual appeal. Considering the huge gap between human and machine visual systems, this paper proposes the topic: Image Quality Assessment for Machine Vision for the first time. Specifically, we (1) defined the subjective preferences of machines, including downstream tasks, test models, and evaluation metrics; (2) established the Machine Preference Database (MPD), which contains 2.25M fine-grained annotations and 30k reference/distorted image pair instances; (3) verified the performance of mainstream IQA algorithms on MPD. Experiments show that current IQA metrics are human-centric and cannot accurately characterize machine preferences. We sincerely hope that MPD can promote the evolution of IQA from human to machine preferences. Project page is on: https://github.com/lcysyzxdxc/MPD.

Zicheng Zhang, Haoning Wu, Ziheng Jia et al.

Image quality scoring and interpreting are two fundamental components of Image Quality Assessment (IQA). The former quantifies image quality, while the latter enables descriptive question answering about image quality. Traditionally, these two tasks have been addressed independently. However, from the perspective of the Human Visual System (HVS) and the Perception-Decision Integration Model, they are inherently interconnected: interpreting serves as the foundation for scoring, while scoring provides an abstract summary of interpreting. Thus, unifying these capabilities within a single model is both intuitive and logically coherent. In this paper, we propose Q-SiT (Quality Scoring and Interpreting joint Teaching), a unified framework that enables large multimodal models (LMMs) to learn both image quality scoring and interpreting simultaneously. We achieve this by transforming conventional IQA datasets into learnable question-answering datasets and incorporating human-annotated quality interpreting data for training. Furthermore, we introduce an efficient scoring & interpreting balance strategy, which first determines the optimal data mix ratio on lightweight LMMs and then maps this ratio to primary LMMs for fine-tuning adjustment. This strategy not only mitigates task interference and enhances cross-task knowledge transfer but also significantly reduces computational costs compared to direct optimization on full-scale LMMs. With this joint learning framework and corresponding training strategy, we develop Q-SiT, the first model capable of simultaneously performing image quality scoring and interpreting tasks, along with its lightweight variant, Q-SiT-mini. Experimental results demonstrate that Q-SiT achieves strong performance in both tasks with superior generalization IQA abilities.Project page at https://github.com/Q-Future/Q-SiT.

Chunyi Li, Xiaozhe Li, Zicheng Zhang et al.

With the emergence of Multimodal Large Language Models (MLLMs), hundreds of benchmarks have been developed to ensure the reliability of MLLMs in downstream tasks. However, the evaluation mechanism itself may not be reliable. For developers of MLLMs, questions remain about which benchmark to use and whether the test results meet their requirements. Therefore, we propose a critical principle of Information Density, which examines how much insight a benchmark can provide for the development of MLLMs. We characterize it from four key dimensions: (1) Fallacy, (2) Difficulty, (3) Redundancy, (4) Diversity. Through a comprehensive analysis of more than 10,000 samples, we measured the information density of 19 MLLM benchmarks. Experiments show that using the latest benchmarks in testing can provide more insight compared to previous ones, but there is still room for improvement in their information density. We hope this principle can promote the development and application of future MLLM benchmarks. Project page: https://github.com/lcysyzxdxc/bench4bench

Jiarui Wang, Huiyu Duan, Ziheng Jia et al.

Recent advancements in large multimodal models (LMMs) have driven substantial progress in both text-to-video (T2V) generation and video-to-text (V2T) interpretation tasks. However, current AI-generated videos (AIGVs) still exhibit limitations in terms of perceptual quality and text-video alignment. Therefore, a reliable and scalable automatic model for AIGV evaluation is desirable, which heavily relies on the scale and quality of human annotations. To this end, we present AIGVE-60K, a comprehensive dataset and benchmark for AI-Generated Video Evaluation, which features (i) comprehensive tasks, encompassing 3,050 extensive prompts across 20 fine-grained task dimensions, (ii) the largest human annotations, including 120K mean-opinion scores (MOSs) and 60K question-answering (QA) pairs annotated on 58,500 videos generated from 30 T2V models, and (iii) bidirectional benchmarking and evaluating for both T2V generation and V2T interpretation capabilities. Based on AIGVE-60K, we propose LOVE, a LMM-based metric for AIGV Evaluation from multiple dimensions including perceptual preference, text-video correspondence, and task-specific accuracy in terms of both instance level and model level. Comprehensive experiments demonstrate that LOVE not only achieves state-of-the-art performance on the AIGVE-60K dataset, but also generalizes effectively to a wide range of other AIGV evaluation benchmarks. These findings highlight the significance of the AIGVE-60K dataset. Database and codes are anonymously available at https://github.com/IntMeGroup/LOVE.

Jiarui Wang, Huiyu Duan, Juntong Wang et al.

With the rapid advancement of generative models, the realism of AI-generated images has significantly improved, posing critical challenges for verifying digital content authenticity. Current deepfake detection methods often depend on datasets with limited generation models and content diversity that fail to keep pace with the evolving complexity and increasing realism of the AI-generated content. Large multimodal models (LMMs), widely adopted in various vision tasks, have demonstrated strong zero-shot capabilities, yet their potential in deepfake detection remains largely unexplored. To bridge this gap, we present \textbf{DFBench}, a large-scale DeepFake Benchmark featuring (i) broad diversity, including 540,000 images across real, AI-edited, and AI-generated content, (ii) latest model, the fake images are generated by 12 state-of-the-art generation models, and (iii) bidirectional benchmarking and evaluating for both the detection accuracy of deepfake detectors and the evasion capability of generative models. Based on DFBench, we propose \textbf{MoA-DF}, Mixture of Agents for DeepFake detection, leveraging a combined probability strategy from multiple LMMs. MoA-DF achieves state-of-the-art performance, further proving the effectiveness of leveraging LMMs for deepfake detection. Database and codes are publicly available at https://github.com/IntMeGroup/DFBench.

Xiaorong Zhu, Ziheng Jia, Jiarui Wang et al.

The rapid evolution of Multi-modality Large Language Models (MLLMs) is driving significant advancements in visual understanding and generation. Nevertheless, a comprehensive assessment of their capabilities, concerning the fine-grained physical principles especially in geometric optics, remains underexplored. To address this gap, we introduce GOBench, the first benchmark to systematically evaluate MLLMs' ability across two tasks: 1) Generating Optically Authentic Imagery and 2) Understanding Underlying Optical Phenomena. We curates high-quality prompts of geometric optical scenarios and use MLLMs to construct GOBench-Gen-1k dataset.We then organize subjective experiments to assess the generated imagery based on Optical Authenticity, Aesthetic Quality, and Instruction Fidelity, revealing MLLMs' generation flaws that violate optical principles. For the understanding task, we apply crafted evaluation instructions to test optical understanding ability of eleven prominent MLLMs. The experimental results demonstrate that current models face significant challenges in both optical generation and understanding. The top-performing generative model, GPT-4o-Image, cannot perfectly complete all generation tasks, and the best-performing MLLM model, Gemini-2.5Pro, attains a mere 37.35\% accuracy in optical understanding. Database and codes are publicly available at https://github.com/aiben-ch/GOBench.

Lancheng Gao, Ziheng Jia, Zixuan Xing et al.

Understanding the multi-dimensional attributes and intensity nuances of image-evoked emotions is pivotal for advancing machine empathy and empowering diverse human-computer interaction applications. However, existing models are still limited to coarse-grained emotion perception or deficient reasoning capabilities. To bridge this gap, we introduce EEmoDB, the largest image-evoked emotion understanding dataset to date. It features $5$ analysis dimensions spanning $5$ distinct task categories, facilitating comprehensive interpretation. Specifically, we compile $1.2M$ question-answering (QA) pairs (EEmoDB-QA) from $125k$ images via automated generation, alongside a $36k$ dataset (EEmoDB-Assess) curated from $25k$ images for fine-grained assessment. Furthermore, we propose EEmo-Logic, an all-in-one multimodal large language model (MLLM) developed via instruction fine-tuning and task-customized group relative preference optimization (GRPO) with novel reward design. Extensive experiments demonstrate that EEmo-Logic achieves robust performance in in-domain and cross-domain datasets, excelling in emotion QA and fine-grained assessment. The code is available at https://anonymous.4open.science/r/EEmoLogic.

Zeyu Zhang, Zijian Chen, Zicheng Zhang et al.

Large Multimodal Models (LMMs) have demonstrated impressive capabilities across a wide range of multimodal tasks, achieving ever-increasing performance on various evaluation benchmarks. However, existing benchmarks are typically static and often overlap with pre-training datasets, leading to fixed complexity constraints and substantial data contamination issues. Meanwhile, manually annotated datasets are labor-intensive, time-consuming, and subject to human bias and inconsistency, leading to reliability and reproducibility issues. To address these problems, we propose a fully dynamic multimodal evaluation framework, named Open-ended Visual Puzzle Generation (OVPG), which aims to generate fresh, diverse, and verifiable evaluation data automatically in puzzle-solving tasks. Specifically, the OVPG pipeline consists of a raw material sampling module, a visual content generation module, and a puzzle rule design module, which ensures that each evaluation instance is primitive, highly randomized, and uniquely solvable, enabling continual adaptation to the evolving capabilities of LMMs. Built upon OVPG, we construct PuzzleBench, a dynamic and scalable benchmark comprising 11,840 VQA samples. It features six carefully designed puzzle tasks targeting three core LMM competencies, visual recognition, logical reasoning, and context understanding. PuzzleBench differs from static benchmarks that quickly become outdated. It enables ongoing dataset refreshing through OVPG and a rich set of open-ended puzzle designs, allowing seamless adaptation to the evolving capabilities of LMMs.

Ziheng Jia, Zicheng Zhang, Jiaying Qian et al.

The advent and proliferation of large multi-modal models (LMMs) have introduced new paradigms to computer vision, transforming various tasks into a unified visual question answering framework. Video Quality Assessment (VQA), a classic field in low-level visual perception, focused initially on quantitative video quality scoring. However, driven by advances in LMMs, it is now progressing toward more holistic visual quality understanding tasks. Recent studies in the image domain have demonstrated that Visual Question Answering (VQA) can markedly enhance low-level visual quality evaluation. Nevertheless, related work has not been explored in the video domain, leaving substantial room for improvement. To address this gap, we introduce the VQA2 Instruction Dataset - the first visual question answering instruction dataset that focuses on video quality assessment. This dataset consists of 3 subsets and covers various video types, containing 157,755 instruction question-answer pairs. Then, leveraging this foundation, we present the VQA2 series models. The VQA2 series models interleave visual and motion tokens to enhance the perception of spatial-temporal quality details in videos. We conduct extensive experiments on video quality scoring and understanding tasks, and results demonstrate that the VQA2series models achieve excellent performance in both tasks. Notably, our final model, the VQA2-Assistant, exceeds the renowned GPT-4o in visual quality understanding tasks while maintaining strong competitiveness in quality scoring tasks. Our work provides a foundation and feasible approach for integrating low-level video quality assessment and understanding with LMMs.

Ziheng Jia, Zicheng Zhang, Zeyu Zhang et al.

The data scaling law has been shown to significantly enhance the performance of large multi-modal models (LMMs) across various downstream tasks. However, in the domain of perceptual video quality assessment (VQA), the potential of scaling law remains unprecedented due to the scarcity of labeled resources and the insufficient scale of datasets. To address this, we propose \textbf{OmniVQA}, an efficient framework designed to efficiently build high-quality, human-in-the-loop VQA multi-modal instruction databases (MIDBs). We then scale up to create \textbf{OmniVQA-Chat-400K}, the largest MIDB in the VQA field concurrently. Our focus is on the technical and aesthetic quality dimensions, with abundant in-context instruction data to provide fine-grained VQA knowledge. Additionally, we have built the \textbf{OmniVQA-MOS-20K} dataset to enhance the model's quantitative quality rating capabilities. We then introduce a \textbf{complementary} training strategy that effectively leverages the knowledge from datasets for quality understanding and quality rating tasks. Furthermore, we propose the \textbf{OmniVQA-FG (fine-grain)-Benchmark} to evaluate the fine-grained performance of the models. Our results demonstrate that our models achieve state-of-the-art performance in both quality understanding and rating tasks.

Ziheng Jia, Linhan Cao, Jinliang Han et al.

Developing a robust visual quality assessment (VQualA) large multi-modal model (LMM) requires achieving versatility, powerfulness, and transferability. However, existing VQualA LMMs typically focus on a single task and rely on full-parameter fine-tuning, which makes them prone to overfitting on specific modalities or task types, thereby limiting their generalization capacity and transferability. To address this, we propose a vision-encoder-centered generative pre-training pipeline and develop the VITAL-Series LMMs. (1) We adopt a machine-executed annotation-scrutiny paradigm, constructing over 4.5M vision-language (VL) pairs-the largest VQualA training dataset to date. (2) We employ a multi-task training workflow that simultaneously enhances the model's quantitative scoring precision and strengthens its capability for quality interpretation across both image and video modalities. (3) Building upon the vision encoder, we realize an efficient model zoo extension: the model zoo exhibits strong zero-shot performance, and each paired decoder requires only a swift warm-up using less than 1/1000 of the pre-training data to achieve performance comparable to the fully trained counterpart. Overall, our work lays a cornerstone for advancing toward the foundation LMM for VQualA.

Juntong Wang, Huiyu Duan, Jiarui Wang et al.

With the rapid advancement of large multimodal models (LMMs), recent text-to-image (T2I) models can generate high-quality images and demonstrate great alignment to short prompts. However, they still struggle to effectively understand and follow long and detailed prompts, displaying inconsistent generation. To address this challenge, we introduce LPG-Bench, a comprehensive benchmark for evaluating long-prompt-based text-to-image generation. LPG-Bench features 200 meticulously crafted prompts with an average length of over 250 words, approaching the input capacity of several leading commercial models. Using these prompts, we generate 2,600 images from 13 state-of-the-art models and further perform comprehensive human-ranked annotations. Based on LPG-Bench, we observe that state-of-the-art T2I alignment evaluation metrics exhibit poor consistency with human preferences on long-prompt-based image generation. To address the gap, we introduce a novel zero-shot metric based on text-to-image-to-text consistency, termed TIT, for evaluating long-prompt-generated images. The core concept of TIT is to quantify T2I alignment by directly comparing the consistency between the raw prompt and the LMM-produced description on the generated image, which includes an efficient score-based instantiation TIT-Score and a large-language-model (LLM) based instantiation TIT-Score-LLM. Extensive experiments demonstrate that our framework achieves superior alignment with human judgment compared to CLIP-score, LMM-score, etc., with TIT-Score-LLM attaining a 7.31% absolute improvement in pairwise accuracy over the strongest baseline. LPG-Bench and TIT methods together offer a deeper perspective to benchmark and foster the development of T2I models. All resources will be made publicly available.

Ziheng Jia, Jiaying Qian, Zicheng Zhang et al.

Reinforcement fine-tuning (RFT) is a proliferating paradigm for LMM training. Analogous to high-level reasoning tasks, RFT is similarly applicable to low-level vision domains, including image quality assessment (IQA). Existing RFT-based IQA methods typically use rule-based output rewards to verify the model's rollouts but provide no reward supervision for the "think" process, leaving its correctness and efficacy uncontrolled. Furthermore, these methods typically fine-tune directly on downstream IQA tasks without explicitly enhancing the model's native low-level visual quality perception, which may constrain its performance upper bound. In response to these gaps, we propose the multi-stage RFT IQA framework (Refine-IQA). In Stage-1, we build the Refine-Perception-20K dataset (with 12 main distortions, 20,907 locally-distorted images, and over 55K RFT samples) and design multi-task reward functions to strengthen the model's visual quality perception. In Stage-2, targeting the quality scoring task, we introduce a probability difference reward involved strategy for "think" process supervision. The resulting Refine-IQA Series Models achieve outstanding performance on both perception and scoring tasks-and, notably, our paradigm activates a robust "think" (quality interpreting) capability that also attains exceptional results on the corresponding quality interpreting benchmark.

Jiaying Qian, Ziheng Jia, Zicheng Zhang et al.

The rapid advancement of AI-driven visual generation technologies has catalyzed significant breakthroughs in image manipulation, particularly in achieving photorealistic localized editing effects on natural scene images (NSIs). Despite extensive research on image quality assessment (IQA) for AI-generated images (AGIs), most studies focus on fully AI-generated outputs (e.g., text-to-image generation), leaving the quality assessment of partial-AIGC images (PAIs)-images with localized AI-driven edits an almost unprecedented field. Motivated by this gap, we construct the first large-scale PAI dataset towards explainable partial-AIGC image quality assessment (EPAIQA), the EPAIQA-15K, which includes 15K images with localized AI manipulation in different regions and over 300K multi-dimensional human ratings. Based on this, we leverage large multi-modal models (LMMs) and propose a three-stage model training paradigm. This paradigm progressively trains the LMM for editing region grounding, quantitative quality scoring, and quality explanation. Finally, we develop the EPAIQA series models, which possess explainable quality feedback capabilities. Our work represents a pioneering effort in the perceptual IQA field for comprehensive PAI quality assessment.