Junlin Li

Ren Yang, Radu Timofte, Meisong Zheng et al. · tencent-ai

This paper reviews the NTIRE 2022 Challenge on Super-Resolution and Quality Enhancement of Compressed Video. In this challenge, we proposed the LDV 2.0 dataset, which includes the LDV dataset (240 videos) and 95 additional videos. This challenge includes three tracks. Track 1 aims at enhancing the videos compressed by HEVC at a fixed QP. Track 2 and Track 3 target both the super-resolution and quality enhancement of HEVC compressed video. They require x2 and x4 super-resolution, respectively. The three tracks totally attract more than 600 registrations. In the test phase, 8 teams, 8 teams and 12 teams submitted the final results to Tracks 1, 2 and 3, respectively. The proposed methods and solutions gauge the state-of-the-art of super-resolution and quality enhancement of compressed video. The proposed LDV 2.0 dataset is available at https://github.com/RenYang-home/LDV_dataset. The homepage of this challenge (including open-sourced codes) is at https://github.com/RenYang-home/NTIRE22_VEnh_SR.

Xiang Li, Haoyuan Cao, Shijie Zhao et al. · pku

Video salient object detection (VSOD), as a fundamental computer vision problem, has been extensively discussed in the last decade. However, all existing works focus on addressing the VSOD problem in 2D scenarios. With the rapid development of VR devices, panoramic videos have been a promising alternative to 2D videos to provide immersive feelings of the real world. In this paper, we aim to tackle the video salient object detection problem for panoramic videos, with their corresponding ambisonic audios. A multimodal fusion module equipped with two pseudo-siamese audio-visual context fusion (ACF) blocks is proposed to effectively conduct audio-visual interaction. The ACF block equipped with spherical positional encoding enables the fusion in the 3D context to capture the spatial correspondence between pixels and sound sources from the equirectangular frames and ambisonic audios. Experimental results verify the effectiveness of our proposed components and demonstrate that our method achieves state-of-the-art performance on the ASOD60K dataset.

Eduardo Pérez-Pellitero, Sibi Catley-Chandar, Richard Shaw et al.

This paper reviews the challenge on constrained high dynamic range (HDR) imaging that was part of the New Trends in Image Restoration and Enhancement (NTIRE) workshop, held in conjunction with CVPR 2022. This manuscript focuses on the competition set-up, datasets, the proposed methods and their results. The challenge aims at estimating an HDR image from multiple respective low dynamic range (LDR) observations, which might suffer from under- or over-exposed regions and different sources of noise. The challenge is composed of two tracks with an emphasis on fidelity and complexity constraints: In Track 1, participants are asked to optimize objective fidelity scores while imposing a low-complexity constraint (i.e. solutions can not exceed a given number of operations). In Track 2, participants are asked to minimize the complexity of their solutions while imposing a constraint on fidelity scores (i.e. solutions are required to obtain a higher fidelity score than the prescribed baseline). Both tracks use the same data and metrics: Fidelity is measured by means of PSNR with respect to a ground-truth HDR image (computed both directly and with a canonical tonemapping operation), while complexity metrics include the number of Multiply-Accumulate (MAC) operations and runtime (in seconds).

Xiaopeng Sun, Weiqi Li, Zhenyu Zhang et al.

360° omnidirectional images have gained research attention due to their immersive and interactive experience, particularly in AR/VR applications. However, they suffer from lower angular resolution due to being captured by fisheye lenses with the same sensor size for capturing planar images. To solve the above issues, we propose a two-stage framework for 360° omnidirectional image superresolution. The first stage employs two branches: model A, which incorporates omnidirectional position-aware deformable blocks (OPDB) and Fourier upsampling, and model B, which adds a spatial frequency fusion module (SFF) to model A. Model A aims to enhance the feature extraction ability of 360° image positional information, while Model B further focuses on the high-frequency information of 360° images. The second stage performs same-resolution enhancement based on the structure of model A with a pixel unshuffle operation. In addition, we collected data from YouTube to improve the fitting ability of the transformer, and created pseudo low-resolution images using a degradation network. Our proposed method achieves superior performance and wins the NTIRE 2023 challenge of 360° omnidirectional image super-resolution.

Jinhai Yang, Mengxi Guo, Shijie Zhao et al.

High-resolution (HR) images are usually downscaled to low-resolution (LR) ones for better display and afterward upscaled back to the original size to recover details. Recent work in image rescaling formulates downscaling and upscaling as a unified task and learns a bijective mapping between HR and LR via invertible networks. However, in real-world applications (e.g., social media), most images are compressed for transmission. Lossy compression will lead to irreversible information loss on LR images, hence damaging the inverse upscaling procedure and degrading the reconstruction accuracy. In this paper, we propose the Self-Asymmetric Invertible Network (SAIN) for compression-aware image rescaling. To tackle the distribution shift, we first develop an end-to-end asymmetric framework with two separate bijective mappings for high-quality and compressed LR images, respectively. Then, based on empirical analysis of this framework, we model the distribution of the lost information (including downscaling and compression) using isotropic Gaussian mixtures and propose the Enhanced Invertible Block to derive high-quality/compressed LR images in one forward pass. Besides, we design a set of losses to regularize the learned LR images and enhance the invertibility. Extensive experiments demonstrate the consistent improvements of SAIN across various image rescaling datasets in terms of both quantitative and qualitative evaluation under standard image compression formats (i.e., JPEG and WebP).

Mengxi Guo, Shijie Zhao, Hao Jiang et al.

Most video platforms provide video streaming services with different qualities, and the quality of the services is usually adjusted by the resolution of the videos. So high-resolution videos need to be downsampled for compression. In order to solve the problem of video coding at different resolutions, we propose a rate-guided arbitrary rescaling network (RARN) for video resizing before encoding. To help the RARN be compatible with standard codecs and generate compression-friendly results, an iteratively optimized transformer-based virtual codec (TVC) is introduced to simulate the key components of video encoding and perform bitrate estimation. By iteratively training the TVC and the RARN, we achieved 5%-29% BD-Rate reduction anchored by linear interpolation under different encoding configurations and resolutions, exceeding the previous methods on most test videos. Furthermore, the lightweight RARN structure can process FHD (1080p) content at real-time speed (91 FPS) and obtain a considerable rate reduction.

Weiqi Li, Xuanyu Zhang, Bin Chen et al.

Image quality assessment (IQA) and image restoration are fundamental problems in low-level vision. Although IQA and restoration are closely connected conceptually, most existing work treats them in isolation. Recent advances in unified multimodal understanding-generation models demonstrate promising results and indicate that stronger understanding can improve generative performance. This motivates a single model that unifies IQA and restoration and explicitly studies how IQA can guide restoration, a setting that remains largely underexplored yet highly valuable. In this paper, we propose UARE, to our knowledge the first Unified vision-language model for image quality Assessment, Restoration, and Enhancement. Built on pretrained unified understanding and generation models, we introduce a two-stage training framework. First, a progressive, easy-to-hard schedule expands from single-type distortions to higher-order mixed degradations, enabling UARE to handle multiple degradations. Second, we perform unified fine-tuning of quality understanding and restoration with interleaved text-image data, aligning IQA signals with restoration objectives. Through multi-task co-training, UARE leverages IQA to boost restoration and enhancement performance. Extensive experiments across IQA, restoration, and enhancement tasks demonstrate the effectiveness of UARE. The code and models will be available at https://github.com/lwq20020127/UARE.

Yangneng Chen, Junlin Li, Weijun Yao et al.

Large Vision-Language Models (LVLMs) have achieved remarkable progress in multimodal tasks, yet their reliability is persistently undermined by hallucinations-generating text that contradicts visual input. Recent studies often attribute these errors to inadequate visual attention. In this work, we analyze the attention mechanisms via the logit lens, uncovering a distinct anomaly we term Vocabulary Hijacking. We discover that specific visual tokens, defined as Inert Tokens, disproportionately attract attention. Crucially, when their intermediate hidden states are projected into the vocabulary space, they consistently decode to a fixed set of unrelated words (termed Hijacking Anchors) across layers, revealing a rigid semantic collapse. Leveraging this semantic rigidity, we propose Hijacking Anchor-Based Identification (HABI), a robust strategy to accurately localize these Inert Tokens. To quantify the impact of this phenomenon, we introduce the Non-Hijacked Visual Attention Ratio (NHAR), a novel metric designed to identify attention heads that remain resilient to hijacking and are critical for factual accuracy. Building on these insights, we propose Hijacking-Aware Visual Attention Enhancement (HAVAE), a training-free intervention that selectively strengthens the focus of these identified heads on salient visual content. Extensive experiments across multiple benchmarks demonstrate that HAVAE significantly mitigates hallucinations with no additional computational overhead, while preserving the model's general capabilities. Our code is publicly available at https://github.com/lab-klc/HAVAE.

Weiqi Li, Xuanyu Zhang, Shijie Zhao et al.

Image quality assessment (IQA) focuses on the perceptual visual quality of images, playing a crucial role in downstream tasks such as image reconstruction, compression, and generation. The rapid advancement of multi-modal large language models (MLLMs) has significantly broadened the scope of IQA, moving toward comprehensive image quality understanding that incorporates content analysis, degradation perception, and comparison reasoning beyond mere numerical scoring. Previous MLLM-based methods typically either generate numerical scores lacking interpretability or heavily rely on supervised fine-tuning (SFT) using large-scale annotated datasets to provide descriptive assessments, limiting their flexibility and applicability. In this paper, we propose Q-Insight, a reinforcement learning-based model built upon group relative policy optimization (GRPO), which demonstrates strong visual reasoning capability for image quality understanding while requiring only a limited amount of rating scores and degradation labels. By jointly optimizing score regression and degradation perception tasks with carefully designed reward functions, our approach effectively exploits their mutual benefits for enhanced performance. Extensive experiments demonstrate that Q-Insight substantially outperforms existing state-of-the-art methods in both score regression and degradation perception tasks, while exhibiting impressive zero-shot generalization to comparison reasoning tasks. Code will be available at https://github.com/lwq20020127/Q-Insight.

Kanglei Zhou, Junlin Li, Ruizhi Cai et al.

Action Quality Assessment (AQA) is pivotal for quantifying actions across domains like sports and medical care. Existing methods often rely on pre-trained backbones from large-scale action recognition datasets to boost performance on smaller AQA datasets. However, this common strategy yields suboptimal results due to the inherent struggle of these backbones to capture the subtle cues essential for AQA. Moreover, fine-tuning on smaller datasets risks overfitting. To address these issues, we propose Coarse-to-Fine Instruction Alignment (CoFInAl). Inspired by recent advances in large language model tuning, CoFInAl aligns AQA with broader pre-trained tasks by reformulating it as a coarse-to-fine classification task. Initially, it learns grade prototypes for coarse assessment and then utilizes fixed sub-grade prototypes for fine-grained assessment. This hierarchical approach mirrors the judging process, enhancing interpretability within the AQA framework. Experimental results on two long-term AQA datasets demonstrate CoFInAl achieves state-of-the-art performance with significant correlation gains of 5.49% and 3.55% on Rhythmic Gymnastics and Fis-V, respectively. Our code is available at https://github.com/ZhouKanglei/CoFInAl_AQA.

Mengxi Guo, Shijie Zhao, Junlin Li et al.

Preprocessing is a well-established technique for optimizing compression, yet existing methods are predominantly Rate-Distortion (R-D) optimized and constrained by pixel-level fidelity. This work pioneers a shift towards Rate-Perception (R-P) optimization by, for the first time, adapting a large-scale pre-trained diffusion model for compression preprocessing. We propose a two-stage framework: first, we distill the multi-step Stable Diffusion 2.1 into a compact, one-step image-to-image model using Consistent Score Identity Distillation (CiD). Second, we perform a parameter-efficient fine-tuning of the distilled model's attention modules, guided by a Rate-Perception loss and a differentiable codec surrogate. Our method seamlessly integrates with standard codecs without any modification and leverages the model's powerful generative priors to enhance texture and mitigate artifacts. Experiments show substantial R-P gains, achieving up to a 30.13% BD-rate reduction in DISTS on the Kodak dataset and delivering superior subjective visual quality.

Guodong Du, Zitao Fang, Jing Li et al.

Foundation models and their checkpoints have significantly advanced deep learning, boosting performance across various applications. However, fine-tuned models often struggle outside their specific domains and exhibit considerable redundancy. Recent studies suggest that combining a pruned fine-tuned model with the original pre-trained model can mitigate forgetting, reduce interference when merging model parameters across tasks, and improve compression efficiency. In this context, developing an effective pruning strategy for fine-tuned models is crucial. Leveraging the advantages of the task vector mechanism, we preprocess fine-tuned models by calculating the differences between them and the original model. Recognizing that different task vector subspaces contribute variably to model performance, we introduce a novel method called Neural Parameter Search (NPS-Pruning) for slimming down fine-tuned models. This method enhances pruning efficiency by searching through neural parameters of task vectors within low-rank subspaces. Our method has three key applications: enhancing knowledge transfer through pairwise model interpolation, facilitating effective knowledge fusion via model merging, and enabling the deployment of compressed models that retain near-original performance while significantly reducing storage costs. Extensive experiments across vision, NLP, and multi-modal benchmarks demonstrate the effectiveness and robustness of our approach, resulting in substantial performance gains. The code is publicly available at: https://github.com/duguodong7/NPS-Pruning.

Xuanyu Zhang, Weiqi Li, Qunliang Xing et al.

Video restoration in real-world scenarios is challenged by heterogeneous degradations, where static architectures and fixed inference pipelines often fail to generalize. Recent agent-based approaches offer dynamic decision making, yet existing video restoration agents remain limited by insufficient quality perception and inefficient search strategies. We propose VQ-Jarvis, a retrieval-augmented, all-in-one intelligent video restoration agent with sharper vision and faster thought. VQ-Jarvis is designed to accurately perceive degradations and subtle differences among paired restoration results, while efficiently discovering optimal restoration trajectories. To enable sharp vision, we construct VSR-Compare, the first large-scale video paired enhancement dataset with 20K comparison pairs covering 7 degradation types, 11 enhancement operators, and diverse content domains. Based on this dataset, we train a multiple operator judge model and a degradation perception model to guide agent decisions. To achieve fast thought, we introduce a hierarchical operator scheduling strategy that adapts to video difficulty: for easy cases, optimal restoration trajectories are retrieved in a one-step manner from a retrieval-augmented generation (RAG) library; for harder cases, a step-by-step greedy search is performed to balance efficiency and accuracy. Extensive experiments demonstrate that VQ-Jarvis consistently outperforms existing methods on complex degraded videos.

Junlin Li, Xinhao Song, Siqi Wang et al.

Text-driven motion editing and intra-structural retargeting, where source and target share topology but may differ in bone lengths, are traditionally handled by fragmented pipelines with incompatible inputs and representations: editing relies on specialized generative steering, while retargeting is deferred to geometric post-processing. We present a unifying perspective where both tasks are cast as instances of conditional transport within a single generative framework. By leveraging recent advances in flow matching, we demonstrate that editing and retargeting are fundamentally the same generative task, distinguished only by which conditioning signal, semantic or structural, is modulated during inference. We implement this vision via a rectified-flow motion model jointly conditioned on text prompts and target skeletal structures. Our architecture extends a DiT-style transformer with per-joint tokenization and explicit joint self-attention to strictly enforce kinematic dependencies, while a multi-condition classifier-free guidance strategy balances text adherence with skeletal conformity. Experiments on SnapMoGen and a multi-character Mixamo subset show that a single trained model supports text-to-motion generation, zero-shot editing, and zero-shot intra-structural retargeting. This unified approach simplifies deployment and improves structural consistency compared to task-specific baselines.

Junlin Li, Shuangyong Song, Guodong Du et al.

Supervised Fine-Tuning (SFT) accelerates taskspecific large language models (LLMs) development, but the resulting proliferation of finetuned models incurs substantial memory overhead. Delta compression addresses this by retaining a single pre-trained LLM with multiple compressed delta weights. However, existing methods fail on models fine-tuned with largescale datasets. We find that larger SFT data scale amplifies delta parameter magnitude, singular values, and entropy, exacerbating compression errors. To tackle this, we propose DQRELO (Delta Compression via Quantization and Residual Low-Rank), a novel training- and data-free delta compression method. It combines coarse-grained one-bit quantization to capture the dominant structure of the delta, followed by compensated residual low-rank approximation to recover fine-grained details from the smaller residual error. Experiments on various LLMs spanning dense and MoE architectures across multiple domains under this challenging setting demonstrate that DQRELO outperforms existing methods. Moreover, we establish key design principles for delta compression through extensive empirical analysis, demonstrating how task difficulty, architecture, and layer positioning create predictable patterns that can guide optimal compression strategies in production systems.

Manman Yuan, Weiming Jia, Xiong Luo et al.

The precise detection of mild cognitive impairment (MCI) is of significant importance in preventing the deterioration of patients in a timely manner. Although hypergraphs have enhanced performance by learning and analyzing brain networks, they often only depend on vector distances between features at a single scale to infer interactions. In this paper, we deal with a more arduous challenge, hypergraph modelling with synchronization between brain regions, and design a novel framework, i.e., A Multi-scale Hypergraph Network for MCI Detection via Synchronous and Attentive Fusion (MHSA), to tackle this challenge. Specifically, our approach employs the Phase-Locking Value (PLV) to calculate the phase synchronization relationship in the spectrum domain of regions of interest (ROIs) and designs a multi-scale feature fusion mechanism to integrate dynamic connectivity features of functional magnetic resonance imaging (fMRI) from both the temporal and spectrum domains. To evaluate and optimize the direct contribution of each ROI to phase synchronization in the temporal domain, we structure the PLV coefficients dynamically adjust strategy, and the dynamic hypergraph is modelled based on a comprehensive temporal-spectrum fusion matrix. Experiments on the real-world dataset indicate the effectiveness of our strategy. The code is available at https://github.com/Jia-Weiming/MHSA.

Fei Zhao, Mengxi Guo, Shijie Zhao et al.

In recent years, user generated content (UGC) has become the dominant force in internet traffic. However, UGC videos exhibit a higher degree of variability and diverse characteristics compared to traditional encoding test videos. This variance challenges the effectiveness of data-driven machine learning algorithms for optimizing encoding in the broader context of UGC scenarios. To address this issue, we propose a Tri-Dynamic Preprocessing framework for UGC. Firstly, we employ an adaptive factor to regulate preprocessing intensity. Secondly, an adaptive quantization level is employed to fine-tune the codec simulator. Thirdly, we utilize an adaptive lambda tradeoff to adjust the rate-distortion loss function. Experimental results on large-scale test sets demonstrate that our method attains exceptional performance.

Fei Zhao, Mengxi Guo, Shijie Zhao et al.

There has been a growing trend in compressing and transmitting videos from terminals for machine vision tasks. Nevertheless, most video coding optimization method focus on minimizing distortion according to human perceptual metrics, overlooking the heightened demands posed by machine vision systems. In this paper, we propose a video preprocessing framework tailored for machine vision tasks to address this challenge. The proposed method incorporates a neural preprocessor which retaining crucial information for subsequent tasks, resulting in the boosting of rate-accuracy performance. We further introduce a differentiable virtual codec to provide constraints on rate and distortion during the training stage. We directly apply widely used standard codecs for testing. Therefore, our solution can be easily applied to real-world scenarios. We conducted extensive experiments evaluating our compression method on two typical downstream tasks with various backbone networks. The experimental results indicate that our approach can save over 15% of bitrate compared to using only the standard codec anchor version.

Kanglong Fan, Tianhe Wu, Wen Wen et al.

Reasoning-induced vision-language models (VLMs) advance image quality assessment (IQA) with textual reasoning, yet their scalar scores often lack sensitivity and collapse to a few values, so-called discrete collapse. We introduce ME-IQA, a plug-and-play, test-time memory-enhanced re-ranking framework. It (i) builds a memory bank and retrieves semantically and perceptually aligned neighbors using reasoning summaries, (ii) reframes the VLM as a probabilistic comparator to obtain pairwise preference probabilities and fuse this ordinal evidence with the initial score under Thurstone's Case V model, and (iii) performs gated reflection and consolidates memory to improve future decisions. This yields denser, distortion-sensitive predictions and mitigates discrete collapse. Experiments across multiple IQA benchmarks show consistent gains over strong reasoning-induced VLM baselines, existing non-reasoning IQA methods, and test-time scaling alternatives.

Tao Wu, Li Yang, Gen Zhan et al.

Enhancing the temporal understanding of Multimodal Large Language Models (MLLMs) is essential for advancing long-form video analysis, enabling tasks such as temporal localization, action detection, and time-sensitive question answering. While reinforcement learning (RL) has recently been explored for improving temporal reasoning, existing approaches are often confined to limited task types and data, restricting their generalization across diverse temporal understanding scenarios. To address this challenge, we present TempR1, a temporal-aware multi-task reinforcement learning framework that systematically strengthens MLLMs' temporal comprehension. We curate a multi-task corpus that exposes the model to diverse temporal structures and semantics, and build upon the Group Relative Policy Optimization (GRPO) algorithm to achieve stable and effective cross-task optimization. Specifically, we categorize temporal tasks into three correspondence types between predicted intervals and ground-truth instances, and design tailored localization rewards for each, enabling TempR1 to capture fine-grained temporal dependencies and adapt to different temporal patterns. Extensive experiments demonstrate that TempR1 attains state-of-the-art performance across multiple benchmarks. Moreover, its joint optimization over complementary tasks yields a strong synergistic effect, enhancing both generalization and single-task performance, establishing a scalable and principled paradigm for temporal reasoning in MLLMs.

Yan Wang, Shijie Zhao, Junlin Li et al.

Diffusion models have attained remarkable breakthroughs in the real-world super-resolution (SR) task, albeit at slow inference and high demand on devices. To accelerate inference, recent works like GenDR adopt step distillation to minimize the step number to one. However, the memory boundary still restricts the maximum processing size, necessitating tile-by-tile restoration of high-resolution images. Through profiling the pipeline, we pinpoint that the variational auto-encoder (VAE) is the bottleneck of latency and memory. To completely solve the problem, we leverage pixel-(un)shuffle operations to eliminate the VAE, reversing the latent-based GenDR to pixel-space GenDR-Pix. However, upscale with x8 pixelshuffle may induce artifacts of repeated patterns. To alleviate the distortion, we propose a multi-stage adversarial distillation to progressively remove the encoder and decoder. Specifically, we utilize generative features from the previous stage models to guide adversarial discrimination. Moreover, we propose random padding to augment generative features and avoid discriminator collapse. We also introduce a masked Fourier space loss to penalize the outliers of amplitude. To improve inference performance, we empirically integrate a padding-based self-ensemble with classifier-free guidance to improve inference scaling. Experimental results show that GenDR-Pix performs 2.8x acceleration and 60% memory-saving compared to GenDR with negligible visual degradation, surpassing other one-step diffusion SR. Against all odds, GenDR-Pix can restore 4K image in only 1 second and 6GB.

Guodong Du, Xuanning Zhou, Junlin Li et al.

Cross-capability transfer is a key challenge in large language model (LLM) research, with applications in multi-task integration, model compression, and continual learning. Recent works like FuseLLM and FuseChat have demonstrated the potential of transferring multiple model capabilities to lightweight models, enhancing adaptability and efficiency, which motivates our investigation into more efficient cross-capability transfer methods. However, existing approaches primarily focus on small, homogeneous models, limiting their applicability. For large, heterogeneous models, knowledge distillation with full-parameter fine-tuning often overlooks the student model's intrinsic capacity and risks catastrophic forgetting, while PEFT methods struggle to effectively absorb knowledge from source LLMs. To address these issues, we introduce GraftLLM, a novel method that stores source model capabilities in a target model with SkillPack format. This approach preserves general capabilities, reduces parameter conflicts, and supports forget-free continual learning and model fusion. We employ a module-aware adaptive compression strategy to compress parameter updates, ensuring efficient storage while maintaining task-specific knowledge. The resulting SkillPack serves as a compact and transferable knowledge carrier, ideal for heterogeneous model fusion and continual learning. Experiments across various scenarios demonstrate that GraftLLM outperforms existing techniques in knowledge transfer, knowledge fusion, and forget-free learning, providing a scalable and efficient solution for cross-capability transfer. The code is publicly available at: https://github.com/duguodong7/GraftLLM.

Ren Yang, Radu Timofte, Jing Liu et al.

This paper reviews the first NTIRE challenge on quality enhancement of compressed video, with a focus on the proposed methods and results. In this challenge, the new Large-scale Diverse Video (LDV) dataset is employed. The challenge has three tracks. Tracks 1 and 2 aim at enhancing the videos compressed by HEVC at a fixed QP, while Track 3 is designed for enhancing the videos compressed by x265 at a fixed bit-rate. Besides, the quality enhancement of Tracks 1 and 3 targets at improving the fidelity (PSNR), and Track 2 targets at enhancing the perceptual quality. The three tracks totally attract 482 registrations. In the test phase, 12 teams, 8 teams and 11 teams submitted the final results of Tracks 1, 2 and 3, respectively. The proposed methods and solutions gauge the state-of-the-art of video quality enhancement. The homepage of the challenge: https://github.com/RenYang-home/NTIRE21_VEnh

Yan Wang, Yi Liu, Shijie Zhao et al.

To satisfy the rapidly increasing demands on the large image (2K-8K) super-resolution (SR), prevailing methods follow two independent tracks: 1) accelerate existing networks by content-aware routing, and 2) design better super-resolution networks via token mixer refining. Despite directness, they encounter unavoidable defects (e.g., inflexible route or non-discriminative processing) limiting further improvements of quality-complexity trade-off. To erase the drawbacks, we integrate these schemes by proposing a content-aware mixer (CAMixer), which assigns convolution for simple contexts and additional deformable window-attention for sparse textures. Specifically, the CAMixer uses a learnable predictor to generate multiple bootstraps, including offsets for windows warping, a mask for classifying windows, and convolutional attentions for endowing convolution with the dynamic property, which modulates attention to include more useful textures self-adaptively and improves the representation capability of convolution. We further introduce a global classification loss to improve the accuracy of predictors. By simply stacking CAMixers, we obtain CAMixerSR which achieves superior performance on large-image SR, lightweight SR, and omnidirectional-image SR.

Xiaohong Liu, Xiongkuo Min, Guangtao Zhai et al.

This paper reports on the NTIRE 2024 Quality Assessment of AI-Generated Content Challenge, which will be held in conjunction with the New Trends in Image Restoration and Enhancement Workshop (NTIRE) at CVPR 2024. This challenge is to address a major challenge in the field of image and video processing, namely, Image Quality Assessment (IQA) and Video Quality Assessment (VQA) for AI-Generated Content (AIGC). The challenge is divided into the image track and the video track. The image track uses the AIGIQA-20K, which contains 20,000 AI-Generated Images (AIGIs) generated by 15 popular generative models. The image track has a total of 318 registered participants. A total of 1,646 submissions are received in the development phase, and 221 submissions are received in the test phase. Finally, 16 participating teams submitted their models and fact sheets. The video track uses the T2VQA-DB, which contains 10,000 AI-Generated Videos (AIGVs) generated by 9 popular Text-to-Video (T2V) models. A total of 196 participants have registered in the video track. A total of 991 submissions are received in the development phase, and 185 submissions are received in the test phase. Finally, 12 participating teams submitted their models and fact sheets. Some methods have achieved better results than baseline methods, and the winning methods in both tracks have demonstrated superior prediction performance on AIGC.

Qifan Liang, Junlin Li, Zhen Han et al.

Electrocautery or lasers will inevitably generate surgical smoke, which hinders the visual guidance of laparoscopic videos for surgical procedures. The surgical smoke can be classified into different types based on its motion patterns, leading to distinctive spatio-temporal characteristics across smoky laparoscopic videos. However, existing desmoking methods fail to account for such smoke-type-specific distinctions. Therefore, we propose the first Smoke-Type-Aware Laparoscopic Video Desmoking Network (STANet) by introducing two smoke types: Diffusion Smoke and Ambient Smoke. Specifically, a smoke mask segmentation sub-network is designed to jointly conduct smoke mask and smoke type predictions based on the attention-weighted mask aggregation, while a smokeless video reconstruction sub-network is proposed to perform specially desmoking on smoky features guided by two types of smoke mask. To address the entanglement challenges of two smoke types, we further embed a coarse-to-fine disentanglement module into the mask segmentation sub-network, which yields more accurate disentangled masks through the smoke-type-aware cross attention between non-entangled and entangled regions. In addition, we also construct the first large-scale synthetic video desmoking dataset with smoke type annotations. Extensive experiments demonstrate that our method not only outperforms state-of-the-art approaches in quality evaluations, but also exhibits superior generalization across multiple downstream surgical tasks.

Wen Wen, Mu Li, Yabin Zhang et al.

Blind video quality assessment (BVQA) plays a pivotal role in evaluating and improving the viewing experience of end-users across a wide range of video-based platforms and services. Contemporary deep learning-based models primarily analyze video content in its aggressively subsampled format, while being blind to the impact of the actual spatial resolution and frame rate on video quality. In this paper, we propose a modular BVQA model and a method of training it to improve its modularity. Our model comprises a base quality predictor, a spatial rectifier, and a temporal rectifier, responding to the visual content and distortion, spatial resolution, and frame rate changes on video quality, respectively. During training, spatial and temporal rectifiers are dropped out with some probabilities to render the base quality predictor a standalone BVQA model, which should work better with the rectifiers. Extensive experiments on both professionally-generated content and user-generated content video databases show that our quality model achieves superior or comparable performance to current methods. Additionally, the modularity of our model offers an opportunity to analyze existing video quality databases in terms of their spatial and temporal complexity.

Xuanyu Zhang, Weiqi Li, Shijie Zhao et al.

Recent advances in AI-generated content (AIGC) have led to the emergence of powerful text-to-video generation models. Despite these successes, evaluating the quality of AIGC-generated videos remains challenging due to limited generalization, lack of temporal awareness, heavy reliance on large-scale annotated datasets, and the lack of effective interaction with generation models. Most current approaches rely on supervised finetuning of vision-language models (VLMs), which often require large-scale annotated datasets and tend to decouple understanding and generation. To address these shortcomings, we propose VQ-Insight, a novel reasoning-style VLM framework for AIGC video quality assessment. Our approach features: (1) a progressive video quality learning scheme that combines image quality warm-up, general task-specific temporal learning, and joint optimization with the video generation model; (2) the design of multi-dimension scoring rewards, preference comparison rewards, and temporal modeling rewards to enhance both generalization and specialization in video quality evaluation. Extensive experiments demonstrate that VQ-Insight consistently outperforms state-of-the-art baselines in preference comparison, multi-dimension scoring, and natural video scoring, bringing significant improvements for video generation tasks.

Weiqi Li, Shijie Zhao, Bin Chen et al. · pku

With the advent of virtual reality technology, omnidirectional image (ODI) rescaling techniques are increasingly embraced for reducing transmitted and stored file sizes while preserving high image quality. Despite this progress, current ODI rescaling methods predominantly focus on enhancing the quality of images in equirectangular projection (ERP) format, which overlooks the fact that the content viewed on head mounted displays (HMDs) is actually a rendered viewport instead of an ERP image. In this work, we emphasize that focusing solely on ERP quality results in inferior viewport visual experiences for users. Thus, we propose ResVR, which is the first comprehensive framework for the joint Rescaling and Viewport Rendering of ODIs. ResVR allows obtaining LR ERP images for transmission while rendering high-quality viewports for users to watch on HMDs. In our ResVR, a novel discrete pixel sampling strategy is developed to tackle the complex mapping between the viewport and ERP, enabling end-to-end training of ResVR pipeline. Furthermore, a spherical pixel shape representation technique is innovatively derived from spherical differentiation to significantly improve the visual quality of rendered viewports. Extensive experiments demonstrate that our ResVR outperforms existing methods in viewport rendering tasks across different fields of view, resolutions, and view directions while keeping a low transmission overhead.

Weiqi Li, Shijie Zhao, Chong Mou et al.

As virtual reality gains popularity, the demand for controllable creation of immersive and dynamic omnidirectional videos (ODVs) is increasing. While previous text-to-ODV generation methods achieve impressive results, they struggle with content inaccuracies and inconsistencies due to reliance solely on textual inputs. Although recent motion control techniques provide fine-grained control for video generation, directly applying these methods to ODVs often results in spatial distortion and unsatisfactory performance, especially with complex spherical motions. To tackle these challenges, we propose OmniDrag, the first approach enabling both scene- and object-level motion control for accurate, high-quality omnidirectional image-to-video generation. Building on pretrained video diffusion models, we introduce an omnidirectional control module, which is jointly fine-tuned with temporal attention layers to effectively handle complex spherical motion. In addition, we develop a novel spherical motion estimator that accurately extracts motion-control signals and allows users to perform drag-style ODV generation by simply drawing handle and target points. We also present a new dataset, named Move360, addressing the scarcity of ODV data with large scene and object motions. Experiments demonstrate the significant superiority of OmniDrag in achieving holistic scene-level and fine-grained object-level control for ODV generation. The project page is available at https://lwq20020127.github.io/OmniDrag.

Jinhai Yang, Mengxi Guo, Shijie Zhao et al.

Adaptive video streaming requires efficient bitrate ladder construction to meet heterogeneous network conditions and end-user demands. Per-title optimized encoding typically traverses numerous encoding parameters to search the Pareto-optimal operating points for each video. Recently, researchers have attempted to predict the content-optimized bitrate ladder for pre-encoding overhead reduction. However, existing methods commonly estimate the encoding parameters on the Pareto front and still require subsequent pre-encodings. In this paper, we propose to directly predict the optimal transcoding resolution at each preset bitrate for efficient bitrate ladder construction. We adopt a Temporal Attentive Gated Recurrent Network to capture spatial-temporal features and predict transcoding resolutions as a multi-task classification problem. We demonstrate that content-optimized bitrate ladders can thus be efficiently determined without any pre-encoding. Our method well approximates the ground-truth bitrate-resolution pairs with a slight Bjøntegaard Delta rate loss of 1.21% and significantly outperforms the state-of-the-art fixed ladder.

Yansong Xu, Xiaohui Wang, Junlin Li et al.

The anthropomorphism of grasping process significantly benefits the experience and grasping efficiency of prosthetic hand wearers. Currently, prosthetic hands controlled by signals such as brain-computer interfaces (BCI) and electromyography (EMG) face difficulties in precisely recognizing the amputees' grasping gestures and executing anthropomorphic grasp processes. Although prosthetic hands equipped with vision systems enables the objects' feature recognition, they lack perception of human grasping intention. Therefore, this paper explores the estimation of grasping gestures solely through visual data to accomplish anthropopathic grasping control and the determination of grasping intention within a multi-object environment. To address this, we propose the Spatial Geometry-based Gesture Mapping (SG-GM) method, which constructs gesture functions based on the geometric features of the human hand grasping processes. It's subsequently implemented on the prosthetic hand. Furthermore, we propose the Motion Trajectory Regression-based Grasping Intent Estimation (MTR-GIE) algorithm. This algorithm predicts pre-grasping object utilizing regression prediction and prior spatial segmentation estimation derived from the prosthetic hand's position and trajectory. The experiments were conducted to grasp 8 common daily objects including cup, fork, etc. The experimental results presented a similarity coefficient $R^{2}$ of grasping process of 0.911, a Root Mean Squared Error ($RMSE$) of 2.47\degree, a success rate of grasping of 95.43$\%$, and an average duration of grasping process of 3.07$\pm$0.41 s. Furthermore, grasping experiments in a multi-object environment were conducted. The average accuracy of intent estimation reached 94.35$\%$. Our methodologies offer a groundbreaking approach to enhance the prosthetic hand's functionality and provides valuable insights for future research.

Yingxue Pang, Shijie Zhao, Haiqiang Wang et al.

Sharpening is a widely adopted technique to improve video quality, which can effectively emphasize textures and alleviate blurring. However, increasing the sharpening level comes with a higher video bitrate, resulting in degraded Quality of Service (QoS). Furthermore, the video quality does not necessarily improve with increasing sharpening levels, leading to issues such as over-sharpening. Clearly, it is essential to figure out how to boost video quality with a proper sharpening level while also controlling bandwidth costs effectively. This paper thus proposes a novel Frequency-assisted Sharpening level Prediction model (FreqSP). We first label each video with the sharpening level correlating to the optimal bitrate and quality tradeoff as ground truth. Then taking uncompressed source videos as inputs, the proposed FreqSP leverages intricate CNN features and high-frequency components to estimate the optimal sharpening level. Extensive experiments demonstrate the effectiveness of our method.

Junlin Li, Guodong DU, Jing Li et al.

Fine-tuning Large Language Models (LLMs) with multimodal encoders on modality-specific data expands the modalities that LLMs can handle, leading to the formation of Multimodal LLMs (MLLMs). However, this paradigm heavily relies on resource-intensive and inflexible fine-tuning from scratch with new multimodal data. In this paper, we propose MMER (Multi-modality Expansion and Retention), a training-free approach that integrates existing MLLMs for effective multimodal expansion while retaining their original performance. Specifically, MMER reuses MLLMs' multimodal encoders while merging their LLM parameters. By comparing original and merged LLM parameters, MMER generates binary masks to approximately separate LLM parameters for each modality. These decoupled parameters can independently process modality-specific inputs, reducing parameter conflicts and preserving original MLLMs' fidelity. MMER can also mitigate catastrophic forgetting by applying a similar process to MLLMs fine-tuned on new tasks. Extensive experiments show significant improvements over baselines, proving that MMER effectively expands LLMs' multimodal capabilities while retaining 99% of the original performance, and also markedly mitigates catastrophic forgetting.

Yansong Xu, Junlin Li, Wei Zhang et al.

3D gaussian splatting has advanced simultaneous localization and mapping (SLAM) technology by enabling real-time positioning and the construction of high-fidelity maps. However, the uncertainty in gaussian position and initialization parameters introduces challenges, often requiring extensive iterative convergence and resulting in redundant or insufficient gaussian representations. To address this, we introduce a novel adaptive densification method based on Fourier frequency domain analysis to establish gaussian priors for rapid convergence. Additionally, we propose constructing independent and unified sparse and dense maps, where a sparse map supports efficient tracking via Generalized Iterative Closest Point (GICP) and a dense map creates high-fidelity visual representations. This is the first SLAM system leveraging frequency domain analysis to achieve high-quality gaussian mapping in real-time. Experimental results demonstrate an average frame rate of 36 FPS on Replica and TUM RGB-D datasets, achieving competitive accuracy in both localization and mapping.

Shijie Zhao, Xuanyu Zhang, Bin Chen et al.

Aligning generative real-world image super-resolution models with human visual preference is challenging due to the perception--fidelity trade-off and diverse, unknown degradations. Prior approaches rely on offline preference optimization and static metric aggregation, which are often non-interpretable and prone to pseudo-diversity under strong conditioning. We propose OARS, a process-aware online alignment framework built on COMPASS, a MLLM-based reward that evaluates the LR to SR transition by jointly modeling fidelity preservation and perceptual gain with an input-quality-adaptive trade-off. To train COMPASS, we curate COMPASS-20K spanning synthetic and real degradations, and introduce a three-stage perceptual annotation pipeline that yields calibrated, fine-grained training labels. Guided by COMPASS, OARS performs progressive online alignment from cold-start flow matching to full-reference and finally reference-free RL via shallow LoRA optimization for on-policy exploration. Extensive experiments and user studies demonstrate consistent perceptual improvements while maintaining fidelity, achieving state-of-the-art performance on Real-ISR benchmarks.

Shijie Zhao, Xuanyu Zhang, Weiqi Li et al.

Reasoning-based image quality assessment (IQA) models trained through reinforcement learning (RL) exhibit exceptional generalization, yet the underlying mechanisms and critical factors driving this capability remain underexplored in current research. Moreover, despite their superior performance, these models incur inference energy usage and latency orders of magnitude higher than their earlier counterparts, restricting their deployment in specific scenarios. Through extensive experiments, this paper verifies and elaborates that through RL training, MLLMs leverage their reasoning capability to convert redundant visual representations into compact, cross-domain aligned text representations. This conversion is precisely the source of the generalization exhibited by these reasoning-based IQA models. Building on this fundamental insight, we propose a novel algorithm, RALI, which employs contrastive learning to directly align images with these generalizable text representations learned by RL. This approach eliminates the reliance on reasoning processes and even obviates the need to load an LLM. For the quality scoring task, this framework achieves generalization performance comparable to reasoning-based models while requiring less than 5% of their model parameters and inference time.

Wen Wen, Tianwu Zhi, Kanglong Fan et al.

Improving vision-language models (VLMs) in the post-training stage typically relies on supervised fine-tuning or reinforcement learning, methods that necessitate costly, human-annotated data. While self-supervised techniques such as self-consistency have proven effective for enhancing reasoning capabilities, their application to perceptual domains such as image quality assessment (IQA) remains largely unexplored. In this work, we introduce EvoQuality, a novel framework that enables a VLM to autonomously refine its quality perception capabilities without any ground-truth labels. EvoQuality adapts the principle of self-consistency to the ranking-based nature of IQA. It generates pseudo-labels by performing pairwise majority voting on the VLM's own outputs to establish a consensus on relative quality. These pseudo-rankings are then formulated into a fidelity reward that guides the model's iterative evolution through group relative policy optimization (GRPO). By iteratively leveraging its own predictions, EvoQuality progressively refines the VLM's perceptual capability. Extensive experiments show that EvoQuality boosts the base VLM's zero-shot performance by 31.8\% on PLCC across diverse IQA benchmarks. Remarkably, despite being entirely self-supervised, EvoQuality achieves performance that is competitive with, or even surpasses, state-of-the-art supervised VLM-based IQA models, outperforming these models on 5 out of 7 IQA benchmarks.

Yingxue Pang, Shijie Zhao, Junlin Li et al.

High-frequency components are crucial for maintaining video clarity and realism, but they also significantly impact coding bitrate, resulting in increased bandwidth and storage costs. This paper presents an end-to-end learning-based framework for adaptive high-frequency preprocessing to enhance subjective quality and save bitrate in video coding. The framework employs the Frequency-attentive Feature pyramid Prediction Network (FFPN) to predict the optimal high-frequency preprocessing strategy, guiding subsequent filtering operators to achieve the optimal tradeoff between bitrate and quality after compression. For training FFPN, we pseudo-label each training video with the optimal strategy, determined by comparing the rate-distortion (RD) performance across different preprocessing types and strengths. Distortion is measured using the latest quality assessment metric. Comprehensive evaluations on multiple datasets demonstrate the visually appealing enhancement capabilities and bitrate savings achieved by our framework.

Yingxue Pang, Shijie Zhao, Mengxi Guo et al.

Sharpening is a widely adopted video enhancement technique. However, uniform sharpening intensity ignores texture variations, degrading video quality. Sharpening also increases bitrate, and there's a lack of techniques to optimally allocate these additional bits across diverse regions. Thus, this paper proposes RPO-AdaSharp, an end-to-end region-adaptive video sharpening model for both perceptual enhancement and bitrate savings. We use the coding tree unit (CTU) partition mask as prior information to guide and constrain the allocation of increased bits. Experiments on benchmarks demonstrate the effectiveness of the proposed model qualitatively and quantitatively.

Yan Wang, Shijie Zhao, Kai Chen et al.

Recent research applying text-to-image (T2I) diffusion models to real-world super-resolution (SR) has achieved remarkable success. However, fundamental misalignments between T2I and SR targets result in a dilemma between inference speed and detail fidelity. Specifically, T2I tasks prioritize multi-step inversion to synthesize coherent outputs aligned with textual prompts and shrink the latent space to reduce generating complexity. Contrariwise, SR tasks preserve most information from low-resolution input while solely restoring high-frequency details, thus necessitating sufficient latent space and fewer inference steps. To bridge the gap, we present a one-step diffusion model for generative detail restoration, GenDR, distilled from a tailored diffusion model with larger latent space. In detail, we train a new SD2.1-VAE16 (0.9B) via representation alignment to expand latent space without enlarging the model size. Regarding step-distillation, we propose consistent score identity distillation (CiD) that incorporates SR task-specific loss into score distillation to leverage more SR priors and align the training target. Furthermore, we extend CiD with adversarial learning and representation alignment (CiDA) to enhance perceptual quality and accelerate training. We also polish the pipeline to achieve a more efficient inference. Experimental results demonstrate that GenDR achieves state-of-the-art performance in both quantitative metrics and visual fidelity.

Ming Cheng, Haoyu Ma, Qiufang Ma et al.

Multi-stage strategies are frequently employed in image restoration tasks. While transformer-based methods have exhibited high efficiency in single-image super-resolution tasks, they have not yet shown significant advantages over CNN-based methods in stereo super-resolution tasks. This can be attributed to two key factors: first, current single-image super-resolution transformers are unable to leverage the complementary stereo information during the process; second, the performance of transformers is typically reliant on sufficient data, which is absent in common stereo-image super-resolution algorithms. To address these issues, we propose a Hybrid Transformer and CNN Attention Network (HTCAN), which utilizes a transformer-based network for single-image enhancement and a CNN-based network for stereo information fusion. Furthermore, we employ a multi-patch training strategy and larger window sizes to activate more input pixels for super-resolution. We also revisit other advanced techniques, such as data augmentation, data ensemble, and model ensemble to reduce overfitting and data bias. Finally, our approach achieved a score of 23.90dB and emerged as the winner in Track 1 of the NTIRE 2023 Stereo Image Super-Resolution Challenge.

Jinjin Gu, Haoming Cai, Chao Dong et al.

This paper reports on the NTIRE 2021 challenge on perceptual image quality assessment (IQA), held in conjunction with the New Trends in Image Restoration and Enhancement workshop (NTIRE) workshop at CVPR 2021. As a new type of image processing technology, perceptual image processing algorithms based on Generative Adversarial Networks (GAN) have produced images with more realistic textures. These output images have completely different characteristics from traditional distortions, thus pose a new challenge for IQA methods to evaluate their visual quality. In comparison with previous IQA challenges, the training and testing datasets in this challenge include the outputs of perceptual image processing algorithms and the corresponding subjective scores. Thus they can be used to develop and evaluate IQA methods on GAN-based distortions. The challenge has 270 registered participants in total. In the final testing stage, 13 participating teams submitted their models and fact sheets. Almost all of them have achieved much better results than existing IQA methods, while the winning method can demonstrate state-of-the-art performance.

Ye Yuan, Junlin Li, Liang Li et al.

The study presents a general framework for discovering underlying Partial Differential Equations (PDEs) using measured spatiotemporal data. The method, called Sparse Spatiotemporal System Discovery ($\text{S}^3\text{d}$), decides which physical terms are necessary and which can be removed (because they are physically negligible in the sense that they do not affect the dynamics too much) from a pool of candidate functions. The method is built on the recent development of Sparse Bayesian Learning; which enforces the sparsity in the to-be-identified PDEs, and therefore can balance the model complexity and fitting error with theoretical guarantees. Without leveraging prior knowledge or assumptions in the discovery process, we use an automated approach to discover ten types of PDEs, including the famous Navier-Stokes and sine-Gordon equations, from simulation data alone. Moreover, we demonstrate our data-driven discovery process with the Complex Ginzburg-Landau Equation (CGLE) using data measured from a traveling-wave convection experiment. Our machine discovery approach presents solutions that has the potential to inspire, support and assist physicists for the establishment of physical laws from measured spatiotemporal data, especially in notorious fields that are often too complex to allow a straightforward establishment of physical law, such as biophysics, fluid dynamics, neuroscience or nonlinear optics.