Zheng Liu

Chunran Zheng, Wei Xu, Zuhao Zou et al.

This paper proposes FAST-LIVO2: a fast, direct LiDAR-inertial-visual odometry framework to achieve accurate and robust state estimation in SLAM tasks and provide great potential in real-time, onboard robotic applications. FAST-LIVO2 fuses the IMU, LiDAR and image measurements efficiently through an ESIKF. To address the dimension mismatch between the heterogeneous LiDAR and image measurements, we use a sequential update strategy in the Kalman filter. To enhance the efficiency, we use direct methods for both the visual and LiDAR fusion, where the LiDAR module registers raw points without extracting edge or plane features and the visual module minimizes direct photometric errors without extracting ORB or FAST corner features. The fusion of both visual and LiDAR measurements is based on a single unified voxel map where the LiDAR module constructs the geometric structure for registering new LiDAR scans and the visual module attaches image patches to the LiDAR points. To enhance the accuracy of image alignment, we use plane priors from the LiDAR points in the voxel map (and even refine the plane prior) and update the reference patch dynamically after new images are aligned. Furthermore, to enhance the robustness of image alignment, FAST-LIVO2 employs an on-demanding raycast operation and estimates the image exposure time in real time. Lastly, we detail three applications of FAST-LIVO2: UAV onboard navigation demonstrating the system's computation efficiency for real-time onboard navigation, airborne mapping showcasing the system's mapping accuracy, and 3D model rendering (mesh-based and NeRF-based) underscoring the suitability of our reconstructed dense map for subsequent rendering tasks. We open source our code, dataset and application on GitHub to benefit the robotics community.

Zhixiang Chi, Rasoul Mohammadi Nasiri, Zheng Liu et al.

Video frame interpolation~(VFI) algorithms have improved considerably in recent years due to unprecedented progress in both data-driven algorithms and their implementations. Recent research has introduced advanced motion estimation or novel warping methods as the means to address challenging VFI scenarios. However, none of the published VFI works considers the spatially non-uniform characteristics of the interpolation error (IE). This work introduces such a solution. By closely examining the correlation between optical flow and IE, the paper proposes novel error prediction metrics that partition the middle frame into distinct regions corresponding to different IE levels. Building upon this IE-driven segmentation, and through the use of novel error-controlled loss functions, it introduces an ensemble of spatially adaptive interpolation units that progressively processes and integrates the segmented regions. This spatial ensemble results in an effective and computationally attractive VFI solution. Extensive experimentation on popular video interpolation benchmarks indicates that the proposed solution outperforms the current state-of-the-art (SOTA) in applications of current interest.

Peitian Zhang, Ninglu Shao, Zheng Liu et al.

We extend the context length of Llama-3-8B-Instruct from 8K to 80K via QLoRA fine-tuning. The entire training cycle is super efficient, which takes 8 hours on one 8xA800 (80G) GPU machine. The resulted model exhibits superior performances across a broad range of evaluation tasks, such as NIHS, topic retrieval, and long-context language understanding; meanwhile, it also well preserves the original capability over short contexts. The dramatic context extension is mainly attributed to merely 3.5K synthetic training samples generated by GPT-4 , which indicates the LLMs' inherent (yet largely underestimated) potential to extend its original context length. In fact, the context length could be extended far beyond 80K with more computation resources. Therefore, the team will publicly release the entire resources (including data, model, data generation pipeline, training code) so as to facilitate the future research from the community: \url{https://github.com/FlagOpen/FlagEmbedding}.

Zheng Liu, Mengjie Liu, Jingzhou Chen et al.

We introduce FUSION, a family of multimodal large language models (MLLMs) with a fully vision-language alignment and integration paradigm. Unlike existing methods that primarily rely on late-stage modality interaction during LLM decoding, our approach achieves deep, dynamic integration throughout the entire processing pipeline. To this end, we propose Text-Guided Unified Vision Encoding, incorporating textual information in vision encoding to achieve pixel-level integration. We further design Context-Aware Recursive Alignment Decoding that recursively aggregates visual features conditioned on textual context during decoding, enabling fine-grained, question-level semantic integration. To guide feature mapping and mitigate modality discrepancies, we develop Dual-Supervised Semantic Mapping Loss. Additionally, we construct a Synthesized Language-Driven Question-Answer (QA) dataset through a new data synthesis method, prioritizing high-quality QA pairs to optimize text-guided feature integration. Building on these foundations, we train FUSION at two scales-3B, 8B-and demonstrate that our full-modality integration approach significantly outperforms existing methods with only 630 vision tokens. Notably, FUSION 3B surpasses Cambrian-1 8B and Florence-VL 8B on most benchmarks. FUSION 3B continues to outperform Cambrian-1 8B even when limited to 300 vision tokens. Our ablation studies show that FUSION outperforms LLaVA-NeXT on over half of the benchmarks under same configuration without dynamic resolution, highlighting the effectiveness of our approach. We release our code, model weights, and dataset. https://github.com/starriver030515/FUSION

Xianfeng Song, Yi Zou, Zheng Shi et al.

Feature-based image matching has extensive applications in computer vision. Keypoints detected in images can be naturally represented as graph structures, and Graph Neural Networks (GNNs) have been shown to outperform traditional deep learning techniques. Consequently, the paradigm of image matching via GNNs has gained significant prominence in recent academic research. In this paper, we first introduce an innovative adaptive graph construction method that utilizes a filtering mechanism based on distance and dynamic threshold similarity. This method dynamically adjusts the criteria for incorporating new vertices based on the characteristics of existing vertices, allowing for the construction of more precise and robust graph structures while avoiding redundancy. We further combine the vertex processing capabilities of GNNs with the global awareness capabilities of Transformers to enhance the model's representation of spatial and feature information within graph structures. This hybrid model provides a deeper understanding of the interrelationships between vertices and their contributions to the matching process. Additionally, we employ the Sinkhorn algorithm to iteratively solve for optimal matching results. Finally, we validate our system using extensive image datasets and conduct comprehensive comparative experiments. Experimental results demonstrate that our system achieves an average improvement of 3.8x-40.3x in overall matching performance. Additionally, the number of vertices and edges significantly impacts training efficiency and memory usage; therefore, we employ multi-GPU technology to accelerate the training process. Our code is available at https://github.com/songxf1024/GIMS.

Zheng Liu, Fu Zhang

A local Bundle Adjustment (BA) on a sliding window of keyframes has been widely used in visual SLAM and proved to be very effective in lowering the drift. But in lidar SLAM, BA method is hardly used because the sparse feature points (e.g., edge and plane) make the exact point matching impossible. In this paper, we formulate the lidar BA as minimizing the distance from a feature point to its matched edge or plane. Unlike the visual SLAM (and prior plane adjustment method in lidar SLAM) where the feature has to be co-determined along with the pose, we show that the feature can be analytically solved and removed from the BA, the resultant BA is only dependent on the scan poses. This greatly reduces the optimization scale and allows large-scale dense plane and edge features to be used. To speedup the optimization, we derive the analytical derivatives of the cost function, up to second order, in closed form. Moreover, we propose a novel adaptive voxelization method to search feature correspondence efficiently. The proposed formulations are incorporated into a LOAM back-end for map refinement. Results show that, although as a back-end, the local BA can be solved very efficiently, even in real-time at 10Hz when optimizing 20 scans of point-cloud. The local BA also considerably lowers the LOAM drift. Our implementation of the BA optimization and LOAM are open-sourced to benefit the community.

Zhen Li, Jinglei Yang, Zheng Liu et al.

Recent advances in image super-resolution (SR) explored the power of deep learning to achieve a better reconstruction performance. However, the feedback mechanism, which commonly exists in human visual system, has not been fully exploited in existing deep learning based image SR methods. In this paper, we propose an image super-resolution feedback network (SRFBN) to refine low-level representations with high-level information. Specifically, we use hidden states in an RNN with constraints to achieve such feedback manner. A feedback block is designed to handle the feedback connections and to generate powerful high-level representations. The proposed SRFBN comes with a strong early reconstruction ability and can create the final high-resolution image step by step. In addition, we introduce a curriculum learning strategy to make the network well suitable for more complicated tasks, where the low-resolution images are corrupted by multiple types of degradation. Extensive experimental results demonstrate the superiority of the proposed SRFBN in comparison with the state-of-the-art methods. Code is avaliable at https://github.com/Paper99/SRFBN_CVPR19.

Weizhi Zhong, Huan Yang, Zheng Liu et al.

Personalized text-to-image generation aims to synthesize images of user-provided concepts in diverse contexts. Despite recent progress in multi-concept personalization, most are limited to object concepts and struggle to customize abstract concepts (e.g., pose, lighting). Some methods have begun exploring multi-concept personalization supporting abstract concepts, but they require test-time fine-tuning for each new concept, which is time-consuming and prone to overfitting on limited training images. In this work, we propose a novel tuning-free method for multi-concept personalization that can effectively customize both object and abstract concepts without test-time fine-tuning. Our method builds upon the modulation mechanism in pre-trained Diffusion Transformers (DiTs) model, leveraging the localized and semantically meaningful properties of the modulation space. Specifically, we propose a novel module, Mod-Adapter, to predict concept-specific modulation direction for the modulation process of concept-related text tokens. It introduces vision-language cross-attention for extracting concept visual features, and Mixture-of-Experts (MoE) layers that adaptively map the concept features into the modulation space. Furthermore, to mitigate the training difficulty caused by the large gap between the concept image space and the modulation space, we introduce a VLM-guided pre-training strategy that leverages the strong image understanding capabilities of vision-language models to provide semantic supervision signals. For a comprehensive comparison, we extend a standard benchmark by incorporating abstract concepts. Our method achieves state-of-the-art performance in multi-concept personalization, supported by quantitative, qualitative, and human evaluations.

Wenjun Wang, Jiacheng Lu, Kejia Chen et al.

Graph similarity computation (GSC) aims to quantify the similarity score between two graphs. Although recent GSC methods based on graph neural networks (GNNs) take advantage of intra-graph structures in message passing, few of them fully utilize the structures presented by edges to boost the representation of their connected nodes. Moreover, previous cross-graph node embedding matching lacks the perception of the overall structure of the graph pair, due to the fact that the node representations from GNNs are confined to the intra-graph structure, causing the unreasonable similarity score. Intuitively, the cross-graph structure represented in the assignment graph is helpful to rectify the inappropriate matching. Therefore, we propose a structure-enhanced graph matching network (SEGMN). Equipped with a dual embedding learning module and a structure perception matching module, SEGMN achieves structure enhancement in both embedding learning and cross-graph matching. The dual embedding learning module incorporates adjacent edge representation into each node to achieve a structure-enhanced representation. The structure perception matching module achieves cross-graph structure enhancement through assignment graph convolution. The similarity score of each cross-graph node pair can be rectified by aggregating messages from structurally relevant node pairs. Experimental results on benchmark datasets demonstrate that SEGMN outperforms the state-of-the-art GSC methods in the GED regression task, and the structure perception matching module is plug-and-play, which can further improve the performance of the baselines by up to 25%.

Zhixiang Chi, Rasoul Mohammadi Nasiri, Zheng Liu et al.

Recent advances in high refresh rate displays as well as the increased interest in high rate of slow motion and frame up-conversion fuel the demand for efficient and cost-effective multi-frame video interpolation solutions. To that regard, inserting multiple frames between consecutive video frames are of paramount importance for the consumer electronics industry. State-of-the-art methods are iterative solutions interpolating one frame at the time. They introduce temporal inconsistencies and clearly noticeable visual artifacts. Departing from the state-of-the-art, this work introduces a true multi-frame interpolator. It utilizes a pyramidal style network in the temporal domain to complete the multi-frame interpolation task in one-shot. A novel flow estimation procedure using a relaxed loss function, and an advanced, cubic-based, motion model is also used to further boost interpolation accuracy when complex motion segments are encountered. Results on the Adobe240 dataset show that the proposed method generates visually pleasing, temporally consistent frames, outperforms the current best off-the-shelf method by 1.57db in PSNR with 8 times smaller model and 7.7 times faster. The proposed method can be easily extended to interpolate a large number of new frames while remaining efficient because of the one-shot mechanism.

Zheng Liu, Jinnian Zhang, Varun Jog et al.

Purpose: The purpose of this study is to investigate the robustness of a commonly-used convolutional neural network for image segmentation with respect to visually-subtle adversarial perturbations, and suggest new methods to make these networks more robust to such perturbations. Materials and Methods: In this retrospective study, the accuracy of brain tumor segmentation was studied in subjects with low- and high-grade gliomas. A three-dimensional UNet model was implemented to segment four different MR series (T1-weighted, post-contrast T1-weighted, T2- weighted, and T2-weighted FLAIR) into four pixelwise labels (Gd-enhancing tumor, peritumoral edema, necrotic and non-enhancing tumor, and background). We developed attack strategies based on the Fast Gradient Sign Method (FGSM), iterative FGSM (i-FGSM), and targeted iterative FGSM (ti-FGSM) to produce effective attacks. Additionally, we explored the effectiveness of distillation and adversarial training via data augmentation to counteract adversarial attacks. Robustness was measured by comparing the Dice coefficient for each attack method using Wilcoxon signed-rank tests. Results: Attacks based on FGSM, i-FGSM, and ti-FGSM were effective in significantly reducing the quality of image segmentation with reductions in Dice coefficient by up to 65%. For attack defenses, distillation performed significantly better than adversarial training approaches. However, all defense approaches performed worse compared to unperturbed test images. Conclusion: Segmentation networks can be adversely affected by targeted attacks that introduce visually minor (and potentially undetectable) modifications to existing images. With an increasing interest in applying deep learning techniques to medical imaging data, it is important to quantify the ramifications of adversarial inputs (either intentional or unintentional).

Lihua Jian, Xiaomin Yang, Zheng Liu et al.

In computer vision and image processing tasks, image fusion has evolved into an attractive research field. However, recent existing image fusion methods are mostly built on pixel-level operations, which may produce unacceptable artifacts and are time-consuming. In this paper, a symmetric encoder-decoder with a residual block (SEDR) for infrared and visible image fusion is proposed. For the training stage, the SEDR network is trained with a new dataset to obtain a fixed feature extractor. For the fusion stage, first, the trained model is utilized to extract the intermediate features and compensation features of two source images. Then, extracted intermediate features are used to generate two attention maps, which are multiplied to the input features for refinement. In addition, the compensation features generated by the first two convolutional layers are merged and passed to the corresponding deconvolutional layers. At last, the refined features are fused for decoding to reconstruct the final fused image. Experimental results demonstrate that the proposed fusion method (named as SEDRFuse) outperforms the state-of-the-art fusion methods in terms of both subjective and objective evaluations.

Xin Fu, Chengkai Zhang, Xiang Peng et al.

Pulsed eddy current (PEC) is an effective electromagnetic non-destructive inspection (NDI) technique for metal materials, which has already been widely adopted in detecting cracking and corrosion in some multi-layer structures. Automatically inspecting the defects in these structures would be conducive to further analysis and treatment of them. In this paper, we propose an effective end-to-end model using convolutional neural networks (CNN) to learn effective features from PEC data. Specifically, we construct a multi-task generic model, based on 1D CNN, to predict both the class and depth of flaws simultaneously. Extensive experiments demonstrate our model is capable of handling both classification and regression tasks on PEC data. Our proposed model obtains higher accuracy and lower error compared to other standard methods.

Shuo Liu, Vijay John, Erik Blasch et al.

Context enhancement is critical for night vision (NV) applications, especially for the dark night situation without any artificial lights. In this paper, we present the infrared-to-visual (IR2VI) algorithm, a novel unsupervised thermal-to-visible image translation framework based on generative adversarial networks (GANs). IR2VI is able to learn the intrinsic characteristics from VI images and integrate them into IR images. Since the existing unsupervised GAN-based image translation approaches face several challenges, such as incorrect mapping and lack of fine details, we propose a structure connection module and a region-of-interest (ROI) focal loss method to address the current limitations. Experimental results show the superiority of the IR2VI algorithm over baseline methods.

Shuo Liu, Zheng Liu

Object Detection is critical for automatic military operations. However, the performance of current object detection algorithms is deficient in terms of the requirements in military scenarios. This is mainly because the object presence is hard to detect due to the indistinguishable appearance and dramatic changes of object's size which is determined by the distance to the detection sensors. Recent advances in deep learning have achieved promising results in many challenging tasks. The state-of-the-art in object detection is represented by convolutional neural networks (CNNs), such as the fast R-CNN algorithm. These CNN-based methods improve the detection performance significantly on several public generic object detection datasets. However, their performance on detecting small objects or undistinguishable objects in visible spectrum images is still insufficient. In this study, we propose a novel detection algorithm for military objects by fusing multi-channel CNNs. We combine spatial, temporal and thermal information by generating a three-channel image, and they will be fused as CNN feature maps in an unsupervised manner. The backbone of our object detection framework is from the fast R-CNN algorithm, and we utilize cross-domain transfer learning technique to fine-tune the CNN model on generated multi-channel images. In the experiments, we validated the proposed method with the images from SENSIAC (Military Sensing Information Analysis Centre) database and compared it with the state-of-the-art. The experimental results demonstrated the effectiveness of the proposed method on both accuracy and computational efficiency.