Jiedong Zhuang

Jiedong Zhuang, Jiaqi Hu, Lianrui Mu et al.

CLIP has achieved impressive zero-shot performance after pre-training on a large-scale dataset consisting of paired image-text data. Previous works have utilized CLIP by incorporating manually designed visual prompts like colored circles and blur masks into the images to guide the model's attention, showing enhanced zero-shot performance in downstream tasks. Although these methods have achieved promising results, they inevitably alter the original information of the images, which can lead to failure in specific tasks. We propose a train-free method Foveal-Attention CLIP (FALIP), which adjusts the CLIP's attention by inserting foveal attention masks into the multi-head self-attention module. We demonstrate FALIP effectively boosts CLIP zero-shot performance in tasks such as referring expressions comprehension, image classification, and 3D point cloud recognition. Experimental results further show that FALIP outperforms existing methods on most metrics and can augment current methods to enhance their performance.

Xiaoyu Liang, Jiayuan Yu, Lianrui Mu et al.

Although Visual-Language Models (VLMs) have shown impressive capabilities in tasks like visual question answering and image captioning, they still struggle with hallucinations. Analysis of attention distribution in these models shows that VLMs tend to processing textual tokens rather than visual tokens. This imbalance of attention distribution causes VLMs to favor textual knowledge in the case of multimodal knowledge conflicts, resulting in differences from the image information. In this paper, we propose Re-Balancing Contrastive Decoding (RBD) method, which employs textual and visual branches to recalibrate attention distribution in VLMs. Specifically, the textual branch injects image noise to stimulate the model's dependency on text, thereby reducing textual bias. Concurrently, the visual branch focuses on the selection of significant tokens, refining the attention mechanism to highlight the primary subject. This dual-branch strategy enables the RBD method to diminish textual bias while enhancing visual information. Experimental results demonstrate that our method, RBD, outperforms the existing methods by the CHAIR and POPE metrics, mitigate hallucinations without reducing the model's general capabilities.

Lianrui Mu, Jianhong Bai, Xiaoxuan He et al.

Enhancing the domain generalization performance of Face Anti-Spoofing (FAS) techniques has emerged as a research focus. Existing methods are dedicated to extracting domain-invariant features from various training domains. Despite the promising performance, the extracted features inevitably contain residual style feature bias (e.g., illumination, capture device), resulting in inferior generalization performance. In this paper, we propose an alternative and effective solution, the Textually Guided Domain Generalization (TeG-DG) framework, which can effectively leverage text information for cross-domain alignment. Our core insight is that text, as a more abstract and universal form of expression, can capture the commonalities and essential characteristics across various attacks, bridging the gap between different image domains. Contrary to existing vision-language models, the proposed framework is elaborately designed to enhance the domain generalization ability of the FAS task. Concretely, we first design a Hierarchical Attention Fusion (HAF) module to enable adaptive aggregation of visual features at different levels; Then, a Textual-Enhanced Visual Discriminator (TEVD) is proposed for not only better alignment between the two modalities but also to regularize the classifier with unbiased text features. TeG-DG significantly outperforms previous approaches, especially in situations with extremely limited source domain data (~14% and ~12% improvements on HTER and AUC respectively), showcasing impressive few-shot performance.

Ming Dai, Wenxuan Cheng, Jiedong Zhuang et al.

Recent advances in visual grounding have largely shifted away from traditional proposal-based two-stage frameworks due to their inefficiency and high computational complexity, favoring end-to-end direct reference paradigms. However, these methods rely exclusively on the referred target for supervision, overlooking the potential benefits of prominent prospective targets. Moreover, existing approaches often fail to incorporate multi-granularity discrimination, which is crucial for robust object identification in complex scenarios. To address these limitations, we propose PropVG, an end-to-end proposal-based framework that, to the best of our knowledge, is the first to seamlessly integrate foreground object proposal generation with referential object comprehension without requiring additional detectors. Furthermore, we introduce a Contrastive-based Refer Scoring (CRS) module, which employs contrastive learning at both sentence and word levels to enhance the capability in understanding and distinguishing referred objects. Additionally, we design a Multi-granularity Target Discrimination (MTD) module that fuses object- and semantic-level information to improve the recognition of absent targets. Extensive experiments on gRefCOCO (GREC/GRES), Ref-ZOM, R-RefCOCO, and RefCOCO (REC/RES) benchmarks demonstrate the effectiveness of PropVG. The codes and models are available at https://github.com/Dmmm1997/PropVG.

Jiedong Zhuang, Lu Lu, Ming Dai et al.

Multimodal large language models (MLLMs) are plagued by exorbitant inference costs attributable to the profusion of visual tokens within the vision encoder. The redundant visual tokens engenders a substantial computational load and key-value (KV) cache footprint bottleneck. Existing approaches focus on token-wise optimization, leveraging diverse intricate token pruning techniques to eliminate non-crucial visual tokens. Nevertheless, these methods often unavoidably undermine the integrity of the KV cache, resulting in failures in long-text generation tasks. To this end, we conduct an in-depth investigation towards the attention mechanism of the model from a new perspective, and discern that attention within more than half of all decode layers are semantic similar. Upon this finding, we contend that the attention in certain layers can be streamlined by inheriting the attention from their preceding layers. Consequently, we propose Lazy Attention, an efficient attention mechanism that enables cross-layer sharing of similar attention patterns. It ingeniously reduces layer-wise redundant computation in attention. In Lazy Attention, we develop a novel layer-shared cache, Q Cache, tailored for MLLMs, which facilitates the reuse of queries across adjacent layers. In particular, Q Cache is lightweight and fully compatible with existing inference frameworks, including Flash Attention and KV cache. Additionally, our method is highly flexible as it is orthogonal to existing token-wise techniques and can be deployed independently or combined with token pruning approaches. Empirical evaluations on multiple benchmarks demonstrate that our method can reduce KV cache usage by over 35% and achieve 1.5x throughput improvement, while sacrificing only approximately 1% of performance on various MLLMs. Compared with SOTA token-wise methods, our technique achieves superior accuracy preservation.

Ming Dai, Jianhong Hu, Jiedong Zhuang et al.

Cross-view geo-localization is a task of matching the same geographic image from different views, e.g., unmanned aerial vehicle (UAV) and satellite. The most difficult challenges are the position shift and the uncertainty of distance and scale. Existing methods are mainly aimed at digging for more comprehensive fine-grained information. However, it underestimates the importance of extracting robust feature representation and the impact of feature alignment. The CNN-based methods have achieved great success in cross-view geo-localization. However it still has some limitations, e.g., it can only extract part of the information in the neighborhood and some scale reduction operations will make some fine-grained information lost. In particular, we introduce a simple and efficient transformer-based structure called Feature Segmentation and Region Alignment (FSRA) to enhance the model's ability to understand contextual information as well as to understand the distribution of instances. Without using additional supervisory information, FSRA divides regions based on the heat distribution of the transformer's feature map, and then aligns multiple specific regions in different views one on one. Finally, FSRA integrates each region into a set of feature representations. The difference is that FSRA does not divide regions manually, but automatically based on the heat distribution of the feature map. So that specific instances can still be divided and aligned when there are significant shifts and scale changes in the image. In addition, a multiple sampling strategy is proposed to overcome the disparity in the number of satellite images and that of images from other sources. Experiments show that the proposed method has superior performance and achieves the state-of-the-art in both tasks of drone view target localization and drone navigation. Code will be released at https://github.com/Dmmm1997/FSRA

Ming Dai, Enhui Zheng, Zhenhua Feng et al.

Unmanned Aerial Vehicles (UAVs) rely on satellite systems for stable positioning. However, due to limited satellite coverage or communication disruptions, UAVs may lose signals from satellite-based positioning systems. In such situations, vision-based techniques can serve as an alternative, ensuring the self-positioning capability of UAVs. However, most of the existing datasets are developed for the geo-localization tasks of the objects identified by UAVs, rather than the self-positioning task of UAVs. Furthermore, the current UAV datasets use discrete sampling on synthetic data, such as Google Maps, thereby neglecting the crucial aspects of dense sampling and the uncertainties commonly experienced in real-world scenarios. To address these issues, this paper presents a new dataset, DenseUAV, which is the first publicly available dataset designed for the UAV self-positioning task. DenseUAV adopts dense sampling on UAV images obtained in low-altitude urban settings. In total, over 27K UAV-view and satellite-view images of 14 university campuses are collected and annotated, establishing a new benchmark. In terms of model development, we first verify the superiority of Transformers over CNNs in this task. Then, we incorporate metric learning into representation learning to enhance the discriminative capacity of the model and to lessen the modality discrepancy. Besides, to facilitate joint learning from both perspectives, we propose a mutually supervised learning approach. Last, we enhance the Recall@K metric and introduce a new measurement, SDM@K, to evaluate the performance of a trained model from both the retrieval and localization perspectives simultaneously. As a result, the proposed baseline method achieves a remarkable Recall@1 score of 83.05% and an SDM@1 score of 86.24% on DenseUAV. The dataset and code will be made publicly available on https://github.com/Dmmm1997/DenseUAV.

Jiedong Zhuang, Lu Lu, Ming Dai et al.

Multimodal large language models (MLLMs) enhance their perceptual capabilities by integrating visual and textual information. However, processing the massive number of visual tokens incurs a significant computational cost. Existing analysis of the MLLM attention mechanisms remains shallow, leading to coarse-grain token pruning strategies that fail to effectively balance speed and accuracy. In this paper, we conduct a comprehensive investigation of MLLM attention mechanisms with LLaVA. We find that numerous visual tokens and partial attention computations are redundant during the decoding process. Based on this insight, we propose Spatial-Temporal Visual Token Trimming ($\textbf{ST}^{3}$), a framework designed to accelerate MLLM inference without retraining. $\textbf{ST}^{3}$ consists of two primary components: 1) Progressive Visual Token Pruning (\textbf{PVTP}), which eliminates inattentive visual tokens across layers, and 2) Visual Token Annealing (\textbf{VTA}), which dynamically reduces the number of visual tokens in each layer as the generated tokens grow. Together, these techniques deliver around $\mathbf{2\times}$ faster inference with only about $\mathbf{30\%}$ KV cache memory compared to the original LLaVA, while maintaining consistent performance across various datasets. Crucially, $\textbf{ST}^{3}$ can be seamlessly integrated into existing pre-trained MLLMs, providing a plug-and-play solution for efficient inference.

Rui Hu, Qian He, Gaofeng He et al.

Modeling and producing lifelike clothed human images has attracted researchers' attention from different areas for decades, with the complexity from highly articulated and structured content. Rendering algorithms decompose and simulate the imaging process of a camera, while are limited by the accuracy of modeled variables and the efficiency of computation. Generative models can produce impressively vivid human images, however still lacking in controllability and editability. This paper studies photorealism enhancement of rendered images, leveraging generative power from diffusion models on the controlled basis of rendering. We introduce a novel framework to translate rendered images into their realistic counterparts, which consists of two stages: Domain Knowledge Injection (DKI) and Realistic Image Generation (RIG). In DKI, we adopt positive (real) domain finetuning and negative (rendered) domain embedding to inject knowledge into a pretrained Text-to-image (T2I) diffusion model. In RIG, we generate the realistic image corresponding to the input rendered image, with a Texture-preserving Attention Control (TAC) to preserve fine-grained clothing textures, exploiting the decoupled features encoded in the UNet structure. Additionally, we introduce SynFashion dataset, featuring high-quality digital clothing images with diverse textures. Extensive experimental results demonstrate the superiority and effectiveness of our method in rendered-to-real image translation.