arXiv subjects

Tianrun Chen

Publications and source records attributed to Tianrun Chen.

49 records · Page 3Linked to original sources

arXiv2024

IBD: Alleviating Hallucinations in Large Vision-Language Models via Image-Biased Decoding

Despite achieving rapid developments and with widespread applications, Large Vision-Language Models (LVLMs) confront a serious challenge of being prone to generating hallucinations. An over-reliance on linguistic priors has been identified as a key factor leading to these hallucinations. In this paper, we propose to alleviate this problem by introducing a novel image-biased decoding (IBD) technique. Our method derives the next-token probability distribution by contrasting predictions from a conventional LVLM with those of an image-biased LVLM, thereby amplifying the correct information highly correlated with image content while mitigating the hallucinatory errors caused by excessive dependence on text. We further conduct a comprehensive statistical analysis to validate the reliability of our method, and design an adaptive adjustment strategy to achieve robust and flexible handling under varying conditions. Experimental results across multiple evaluation metrics verify that our method, despite not requiring additional training data and only with a minimal increase in model parameters, can significantly reduce hallucinations in LVLMs and enhance the truthfulness of the generated response.

cs.CV↗

arXiv2024

RESMatch: Referring Expression Segmentation in a Semi-Supervised Manner

Referring expression segmentation (RES), a task that involves localizing specific instance-level objects based on free-form linguistic descriptions, has emerged as a crucial frontier in human-AI interaction. It demands an intricate understanding of both visual and textual contexts and often requires extensive training data. This paper introduces RESMatch, the first semi-supervised learning (SSL) approach for RES, aimed at reducing reliance on exhaustive data annotation. Extensive validation on multiple RES datasets demonstrates that RESMatch significantly outperforms baseline approaches, establishing a new state-of-the-art. Although existing SSL techniques are effective in image segmentation, we find that they fall short in RES. Facing the challenges including the comprehension of free-form linguistic descriptions and the variability in object attributes, RESMatch introduces a trifecta of adaptations: revised strong perturbation, text augmentation, and adjustments for pseudo-label quality and strong-weak supervision. This pioneering work lays the groundwork for future research in semi-supervised learning for referring expression segmentation.

cs.CV↗

arXiv2023

Deep3DSketch: 3D modeling from Free-hand Sketches with View- and Structural-Aware Adversarial Training

This work aims to investigate the problem of 3D modeling using single free-hand sketches, which is one of the most natural ways we humans express ideas. Although sketch-based 3D modeling can drastically make the 3D modeling process more accessible, the sparsity and ambiguity of sketches bring significant challenges for creating high-fidelity 3D models that reflect the creators' ideas. In this work, we propose a view- and structural-aware deep learning approach, \textit{Deep3DSketch}, which tackles the ambiguity and fully uses sparse information of sketches, emphasizing the structural information. Specifically, we introduced random pose sampling on both 3D shapes and 2D silhouettes, and an adversarial training scheme with an effective progressive discriminator to facilitate learning of the shape structures. Extensive experiments demonstrated the effectiveness of our approach, which outperforms existing methods -- with state-of-the-art (SOTA) performance on both synthetic and real datasets.

cs.MM↗

arXiv2023

Deep3DSketch+: Obtaining Customized 3D Model by Single Free-Hand Sketch through Deep Learning

As 3D models become critical in today's manufacturing and product design, conventional 3D modeling approaches based on Computer-Aided Design (CAD) are labor-intensive, time-consuming, and have high demands on the creators. This work aims to introduce an alternative approach to 3D modeling by utilizing free-hand sketches to obtain desired 3D models. We introduce Deep3DSketch+, which is a deep-learning algorithm that takes the input of a single free-hand sketch and produces a complete and high-fidelity model that matches the sketch input. The neural network has view- and structural-awareness enabled by a Shape Discriminator (SD) and a Stroke Enhancement Module (SEM), which overcomes the limitations of sparsity and ambiguity of the sketches. The network design also brings high robustness to partial sketch input in industrial applications.Our approach has undergone extensive experiments, demonstrating its state-of-the-art (SOTA) performance on both synthetic and real-world datasets. These results validate the effectiveness and superiority of our method compared to existing techniques. We have demonstrated the conversion of free-hand sketches into physical 3D objects using additive manufacturing. We believe that our approach has the potential to accelerate product design and democratize customized manufacturing.

cs.MM↗

arXiv2023

Reality3DSketch: Rapid 3D Modeling of Objects from Single Freehand Sketches

The emerging trend of AR/VR places great demands on 3D content. However, most existing software requires expertise and is difficult for novice users to use. In this paper, we aim to create sketch-based modeling tools for user-friendly 3D modeling. We introduce Reality3DSketch with a novel application of an immersive 3D modeling experience, in which a user can capture the surrounding scene using a monocular RGB camera and can draw a single sketch of an object in the real-time reconstructed 3D scene. A 3D object is generated and placed in the desired location, enabled by our novel neural network with the input of a single sketch. Our neural network can predict the pose of a drawing and can turn a single sketch into a 3D model with view and structural awareness, which addresses the challenge of sparse sketch input and view ambiguity. We conducted extensive experiments synthetic and real-world datasets and achieved state-of-the-art (SOTA) results in both sketch view estimation and 3D modeling performance. According to our user study, our method of performing 3D modeling in a scene is $>$5x faster than conventional methods. Users are also more satisfied with the generated 3D model than the results of existing methods.

cs.HC↗

arXiv2023

Deep3DSketch+\+: High-Fidelity 3D Modeling from Single Free-hand Sketches

The rise of AR/VR has led to an increased demand for 3D content. However, the traditional method of creating 3D content using Computer-Aided Design (CAD) is a labor-intensive and skill-demanding process, making it difficult to use for novice users. Sketch-based 3D modeling provides a promising solution by leveraging the intuitive nature of human-computer interaction. However, generating high-quality content that accurately reflects the creator's ideas can be challenging due to the sparsity and ambiguity of sketches. Furthermore, novice users often find it challenging to create accurate drawings from multiple perspectives or follow step-by-step instructions in existing methods. To address this, we introduce a groundbreaking end-to-end approach in our work, enabling 3D modeling from a single free-hand sketch, Deep3DSketch+$\backslash$+. The issue of sparsity and ambiguity using single sketch is resolved in our approach by leveraging the symmetry prior and structural-aware shape discriminator. We conducted comprehensive experiments on diverse datasets, including both synthetic and real data, to validate the efficacy of our approach and demonstrate its state-of-the-art (SOTA) performance. Users are also more satisfied with results generated by our approach according to our user study. We believe our approach has the potential to revolutionize the process of 3D modeling by offering an intuitive and easy-to-use solution for novice users.

cs.HC↗

arXiv2023

Deep3DSketch+: Rapid 3D Modeling from Single Free-hand Sketches

The rapid development of AR/VR brings tremendous demands for 3D content. While the widely-used Computer-Aided Design (CAD) method requires a time-consuming and labor-intensive modeling process, sketch-based 3D modeling offers a potential solution as a natural form of computer-human interaction. However, the sparsity and ambiguity of sketches make it challenging to generate high-fidelity content reflecting creators' ideas. Precise drawing from multiple views or strategic step-by-step drawings is often required to tackle the challenge but is not friendly to novice users. In this work, we introduce a novel end-to-end approach, Deep3DSketch+, which performs 3D modeling using only a single free-hand sketch without inputting multiple sketches or view information. Specifically, we introduce a lightweight generation network for efficient inference in real-time and a structural-aware adversarial training approach with a Stroke Enhancement Module (SEM) to capture the structural information to facilitate learning of the realistic and fine-detailed shape structures for high-fidelity performance. Extensive experiments demonstrated the effectiveness of our approach with the state-of-the-art (SOTA) performance on both synthetic and real datasets.

cs.CV↗

arXiv2023

Learning Gabor Texture Features for Fine-Grained Recognition

Extracting and using class-discriminative features is critical for fine-grained recognition. Existing works have demonstrated the possibility of applying deep CNNs to exploit features that distinguish similar classes. However, CNNs suffer from problems including frequency bias and loss of detailed local information, which restricts the performance of recognizing fine-grained categories. To address the challenge, we propose a novel texture branch as complimentary to the CNN branch for feature extraction. We innovatively utilize Gabor filters as a powerful extractor to exploit texture features, motivated by the capability of Gabor filters in effectively capturing multi-frequency features and detailed local information. We implement several designs to enhance the effectiveness of Gabor filters, including imposing constraints on parameter values and developing a learning method to determine the optimal parameters. Moreover, we introduce a statistical feature extractor to utilize informative statistical information from the signals captured by Gabor filters, and a gate selection mechanism to enable efficient computation by only considering qualified regions as input for texture extraction. Through the integration of features from the Gabor-filter-based texture branch and CNN-based semantic branch, we achieve comprehensive information extraction. We demonstrate the efficacy of our method on multiple datasets, including CUB-200-2011, NA-bird, Stanford Dogs, and GTOS-mobile. State-of-the-art performance is achieved using our approach.

cs.CV↗

arXiv2023

SAM Fails to Segment Anything? -- SAM-Adapter: Adapting SAM in Underperformed Scenes: Camouflage, Shadow, Medical Image Segmentation, and More

The emergence of large models, also known as foundation models, has brought significant advancements to AI research. One such model is Segment Anything (SAM), which is designed for image segmentation tasks. However, as with other foundation models, our experimental findings suggest that SAM may fail or perform poorly in certain segmentation tasks, such as shadow detection and camouflaged object detection (concealed object detection). This study first paves the way for applying the large pre-trained image segmentation model SAM to these downstream tasks, even in situations where SAM performs poorly. Rather than fine-tuning the SAM network, we propose \textbf{SAM-Adapter}, which incorporates domain-specific information or visual prompts into the segmentation network by using simple yet effective adapters. By integrating task-specific knowledge with general knowledge learnt by the large model, SAM-Adapter can significantly elevate the performance of SAM in challenging tasks as shown in extensive experiments. We can even outperform task-specific network models and achieve state-of-the-art performance in the task we tested: camouflaged object detection, shadow detection. We also tested polyp segmentation (medical image segmentation) and achieves better results. We believe our work opens up opportunities for utilizing SAM in downstream tasks, with potential applications in various fields, including medical image processing, agriculture, remote sensing, and more.

cs.CV↗

arXiv2023

Continual Semantic Segmentation with Automatic Memory Sample Selection

Continual Semantic Segmentation (CSS) extends static semantic segmentation by incrementally introducing new classes for training. To alleviate the catastrophic forgetting issue in CSS, a memory buffer that stores a small number of samples from the previous classes is constructed for replay. However, existing methods select the memory samples either randomly or based on a single-factor-driven handcrafted strategy, which has no guarantee to be optimal. In this work, we propose a novel memory sample selection mechanism that selects informative samples for effective replay in a fully automatic way by considering comprehensive factors including sample diversity and class performance. Our mechanism regards the selection operation as a decision-making process and learns an optimal selection policy that directly maximizes the validation performance on a reward set. To facilitate the selection decision, we design a novel state representation and a dual-stage action space. Our extensive experiments on Pascal-VOC 2012 and ADE 20K datasets demonstrate the effectiveness of our approach with state-of-the-art (SOTA) performance achieved, outperforming the second-place one by 12.54% for the 6stage setting on Pascal-VOC 2012.

cs.CV↗

arXiv2022

Panoptic NeRF: 3D-to-2D Label Transfer for Panoptic Urban Scene Segmentation

Large-scale training data with high-quality annotations is critical for training semantic and instance segmentation models. Unfortunately, pixel-wise annotation is labor-intensive and costly, raising the demand for more efficient labeling strategies. In this work, we present a novel 3D-to-2D label transfer method, Panoptic NeRF, which aims for obtaining per-pixel 2D semantic and instance labels from easy-to-obtain coarse 3D bounding primitives. Our method utilizes NeRF as a differentiable tool to unify coarse 3D annotations and 2D semantic cues transferred from existing datasets. We demonstrate that this combination allows for improved geometry guided by semantic information, enabling rendering of accurate semantic maps across multiple views. Furthermore, this fusion process resolves label ambiguity of the coarse 3D annotations and filters noise in the 2D predictions. By inferring in 3D space and rendering to 2D labels, our 2D semantic and instance labels are multi-view consistent by design. Experimental results show that Panoptic NeRF outperforms existing label transfer methods in terms of accuracy and multi-view consistency on challenging urban scenes of the KITTI-360 dataset.

cs.CV↗

arXiv2021

Recognizing three-dimensional phase images with deep learning

Optical phase contains key information for biomedical and astronomical imaging. However, it is often obscured by layers of heterogeneous and scattering media, which render optical phase imaging at different depths an utmost challenge. Limited by the memory effect, current methods for phase imaging in strong scattering media are inapplicable to retrieving phases at different depths. To address this challenge, we developed a speckle three-dimensional reconstruction network (STRN) to recognize phase objects behind scattering media, which circumvents the limitations of memory effect. From the single-shot, reference-free and scanning-free speckle pattern input, STRN distinguishes depth-resolving quantitative phase information with high fidelity. Our results promise broad applications in biomedical tomography and endoscopy.

physics.optics↗

arXiv2020

Second Harmonic Imaging Enhanced by Deep Learning Decipher

Wavefront sensing and reconstruction are widely used for adaptive optics, aberration correction, and high-resolution optical phase imaging. Traditionally, interference and/or microlens arrays are used to convert the optical phase into intensity variation. Direct imaging of distorted wavefront usually results in complicated phase retrieval with low contrast and low sensitivity. Here, a novel approach has been developed and experimentally demonstrated based on the phase-sensitive information encoded into second harmonic signals, which are intrinsically sensitive to wavefront modulations. By designing and implementing a deep neural network, we demonstrate the second harmonic imaging enhanced by deep learning decipher (SHIELD) for efficient and resilient phase retrieval. Inheriting the advantages of two-photon microscopy, SHIELD demonstrates single-shot, reference-free, and video-rate phase imaging with sensitivity better than λ/100 and high robustness against noises, facilitating numerous applications from biological imaging to wavefront sensing.

physics.optics↗