no code implementations • 12 Apr 2024 • Hai Nguyen-Truong, E-Ro Nguyen, Tuan-Anh Vu, Minh-Triet Tran, Binh-Son Hua, Sai-Kit Yeung
Our method involves using CLIP to derive a CLIP Prior that integrates an object-centric visual heatmap with text description, which can be used as the initial query in DETR-based architecture for the segmentation task.
no code implementations • 25 Jan 2024 • Quang-Trung Truong, Duc Thanh Nguyen, Binh-Son Hua, Sai-Kit Yeung
This is enabled by deformable attention mechanism, where the keys and values capturing the memory of a video sequence in the attention module have flexible locations updated across frames.
Ranked #8 on Unsupervised Video Object Segmentation on DAVIS 2016 val (using extra training data)
no code implementations • 29 Dec 2023 • Tuan-Anh Vu, Duc Thanh Nguyen, Qing Guo, Binh-Son Hua, Nhat Minh Chung, Ivor W. Tsang, Sai-Kit Yeung
Such cross-domain representations are desirable in segmenting camouflaged objects where visual cues are subtle to distinguish the objects from the background, especially in segmenting novel objects which are not seen in training.
no code implementations • 2 Dec 2023 • Uy Dieu Tran, Minh Luu, Phong Nguyen, Janne Heikkila, Khoi Nguyen, Binh-Son Hua
Text-to-3D synthesis has recently emerged as a new approach to sampling 3D models by adopting pretrained text-to-image models as guiding visual priors.
1 code implementation • 30 Nov 2023 • Yingshu Chen, Guocheng Shao, Ka Chun Shum, Binh-Son Hua, Sai-Kit Yeung
Building on such taxonomy, our survey first revisits the background of neural stylization on 2D images, and then provides in-depth discussions on recent neural stylization methods for 3D data, where we also provide a mini-benchmark on artistic stylization methods.
no code implementations • 22 Nov 2023 • Tuan-Anh Vu, Srinjay Sarkar, Zhiyuan Zhang, Binh-Son Hua, Sai-Kit Yeung
We are inspired by the recent revolution of learning implicit representation and point cloud upsampling, which can produce high-quality 3D surface reconstruction and proximity-to-surface, respectively.
1 code implementation • 20 Sep 2023 • Ka Chun Shum, Jaeyeon Kim, Binh-Son Hua, Duc Thanh Nguyen, Sai-Kit Yeung
Specifically, to insert a new foreground object represented by a set of multi-view images into a background radiance field, we use a text-to-image diffusion model to learn and generate combined images that fuse the object of interest into the given background across views.
1 code implementation • ICCV 2023 • Hong-Wing Pang, Binh-Son Hua, Sai-Kit Yeung
In this work, we propose a stylization framework for NeRF based on local style transfer.
1 code implementation • ICCV 2023 • Tuan Duc Ngo, Binh-Son Hua, Khoi Nguyen
Furthermore, we demonstrate the robustness of our approach, where we can adapt various state-of-the-art fully supervised methods to the weak supervision task by using our pseudo labels for training.
no code implementations • ICCV 2023 • Ka Chun Shum, Hong-Wing Pang, Binh-Son Hua, Duc Thanh Nguyen, Sai-Kit Yeung
We use this object layout to condition a generative adversarial network to synthesize images of an input scene.
2 code implementations • CVPR 2023 • Tuan Duc Ngo, Binh-Son Hua, Khoi Nguyen
Existing 3D instance segmentation methods are predominated by the bottom-up design -- manually fine-tuned algorithm to group points into clusters followed by a refinement network.
Ranked #2 on 3D Instance Segmentation on STPLS3D
no code implementations • 16 Nov 2022 • Jaeyeon Kim, Binh-Son Hua, Duc Thanh Nguyen, Sai-Kit Yeung
In this paper, we propose a new method for mapping a 3D point cloud to the latent space of a 3D generative adversarial network.
1 code implementation • 28 Oct 2022 • Bach Tran, Binh-Son Hua, Anh Tuan Tran, Minh Hoai
Inspired by the success of deep learning in the image domain, we devise a novel pre-training technique for better model initialization by utilizing the multi-view rendering of the 3D data.
1 code implementation • 28 Oct 2022 • Phuoc-Hieu Le, Quynh Le, Rang Nguyen, Binh-Son Hua
In this work, we propose a weakly supervised learning method that inverts the physical image formation process for HDR reconstruction via learning to generate multiple exposures from a single image.
1 code implementation • 13 Sep 2022 • Yingshu Chen, Tuan-Anh Vu, Ka-Chun Shum, Binh-Son Hua, Sai-Kit Yeung
Architectural photography is a genre of photography that focuses on capturing a building or structure in the foreground with dramatic lighting in the background.
1 code implementation • 21 Jul 2022 • Khoi D. Nguyen, Quoc-Huy Tran, Khoi Nguyen, Binh-Son Hua, Rang Nguyen
To the best of our knowledge, our work is the first to explore transductive few-shot video classification.
1 code implementation • NeurIPS 2021 • Duong H. Le, Khoi D. Nguyen, Khoi Nguyen, Quoc-Huy Tran, Rang Nguyen, Binh-Son Hua
In this work, we propose to use out-of-distribution samples, i. e., unlabeled samples coming from outside the target classes, to improve few-shot learning.
1 code implementation • 30 Mar 2022 • Tuan-Anh Vu, Duc Thanh Nguyen, Binh-Son Hua, Quang-Hieu Pham, Sai-Kit Yeung
The key insight is simultaneously performing both tasks via learning of spatial and temporal features from a sequence of point clouds can leverage individual tasks, leading to improved overall performance.
Ranked #1 on 3D Human Reconstruction on Dynamic FAUST
1 code implementation • 16 Mar 2022 • Ngoc-Vuong Ho, Tan Nguyen, Gia-Han Diep, Ngan Le, Binh-Son Hua
In this paper, we propose Point-Unet, a novel method that incorporates the efficiency of deep learning with 3D point clouds into volumetric segmentation.
2 code implementations • 16 Mar 2022 • Tan Nguyen, Binh-Son Hua, Ngan Le
Medical image segmentation has been so far achieving promising results with Convolutional Neural Networks (CNNs).
1 code implementation • 26 Feb 2022 • Zhiyuan Zhang, Binh-Son Hua, Sai-Kit Yeung
3D point clouds deep learning is a promising field of research that allows a neural network to learn features of point clouds directly, making it a robust tool for solving 3D scene understanding tasks.
no code implementations • 15 Jan 2022 • Minh Tran, Loi Ly, Binh-Son Hua, Ngan Le
Capsule network is a recent new deep network architecture that has been applied successfully for medical image segmentation tasks.
1 code implementation • 2 Dec 2021 • Tan M. Dinh, Rang Nguyen, Binh-Son Hua
Our study outlines several issues in the current evaluation pipeline: (i) for image quality assessment, a commonly used metric, e. g., Inception Score (IS), is often either miscalibrated for the single-object case or misused for the multi-object case; (ii) for text relevance and object accuracy assessment, there is an overfitting phenomenon in the existing R-precision (RP) and Semantic Object Accuracy (SOA) metrics, respectively; (iii) for multi-object case, many vital factors for evaluation, e. g., object fidelity, positional alignment, counting alignment, are largely dismissed; (iv) the ranking of the methods based on current metrics is highly inconsistent with real images.
1 code implementation • Advances in Neural Information Processing Systems 2021 • Duong H. Le*, Khoi D. Nguyen*, Khoi Nguyen, Quoc-Huy Tran, Rang Nguyen, Binh-Son Hua
In this work, we propose to use out-of-distribution samples, i. e., unlabeled samples coming from outside the target classes, to improve few-shot learning.
1 code implementation • CVPR 2022 • Tan M. Dinh, Anh Tuan Tran, Rang Nguyen, Binh-Son Hua
In the first phase, we train an encoder to map the input image to StyleGAN2 $\mathcal{W}$-space, which was proven to have excellent editability but lower reconstruction quality.
1 code implementation • 4 Aug 2021 • Hong-Wing Pang, Yingshu Chen, Phuoc-Hieu Le, Binh-Son Hua, Duc Thanh Nguyen, Sai-Kit Yeung
In this paper, we introduce a new problem of domain-specific indoor scene image synthesis, namely neural scene decoration.
1 code implementation • ICLR 2021 • Duong H. Le, Binh-Son Hua
Our results emphasize the cruciality of the learning rate schedule in pruned network retraining - a detail often overlooked by practitioners during the implementation of network pruning.
Ranked #10 on Network Pruning on ImageNet
1 code implementation • ICCV 2021 • Trung Nguyen, Quang-Hieu Pham, Tam Le, Tung Pham, Nhat Ho, Binh-Son Hua
From this study, we propose to use sliced Wasserstein distance and its variants for learning representations of 3D point clouds.
no code implementations • ICCV 2021 • Jaeyeon Kim, Binh-Son Hua, Duc Thanh Nguyen, Sai-Kit Yeung
With recent developments of convolutional neural networks, deep learning for 3D point clouds has shown significant progress in various 3D scene understanding tasks, e. g., object recognition, semantic segmentation.
no code implementations • 7 Aug 2020 • Zhiyuan Zhang, Binh-Son Hua, Wei Chen, Yibin Tian, Sai-Kit Yeung
We found that a key reason is that compared to point coordinates, rotation-invariant features consumed by point cloud convolution are not as distinctive.
1 code implementation • 21 Nov 2019 • Quang-Hieu Pham, Mikaela Angelina Uy, Binh-Son Hua, Duc Thanh Nguyen, Gemma Roig, Sai-Kit Yeung
In this work, we present a novel method to learn a local cross-domain descriptor for 2D image and 3D point cloud matching.
1 code implementation • 17 Aug 2019 • Zhiyuan Zhang, Binh-Son Hua, David W. Rosen, Sai-Kit Yeung
Our core idea is to use low-level rotation invariant geometric features such as distances and angles to design a convolution operator for point cloud learning.
1 code implementation • ICCV 2019 • Zhiyuan Zhang, Binh-Son Hua, Sai-Kit Yeung
Deep learning with 3D data has progressed significantly since the introduction of convolutional neural networks that can handle point order ambiguity in point cloud data.
Ranked #8 on 3D Semantic Segmentation on DALES
1 code implementation • ICCV 2019 • Mikaela Angelina Uy, Quang-Hieu Pham, Binh-Son Hua, Duc Thanh Nguyen, Sai-Kit Yeung
From our comprehensive benchmark, we show that our dataset poses great challenges to existing point cloud classification techniques as objects from real-world scans are often cluttered with background and/or are partial due to occlusions.
1 code implementation • CVPR 2019 • Quang-Hieu Pham, Duc Thanh Nguyen, Binh-Son Hua, Gemma Roig, Sai-Kit Yeung
Deep learning techniques have become the to-go models for most vision-related tasks on 2D images.
Ranked #2 on 3D Instance Segmentation on SceneNN
3D Instance Segmentation 3D Semantic Instance Segmentation +3
no code implementations • 1 Apr 2018 • Quang-Hieu Pham, Binh-Son Hua, Duc Thanh Nguyen, Sai-Kit Yeung
The widespread adoption of autonomous systems such as drones and assistant robots has created a need for real-time high-quality semantic scene segmentation.
1 code implementation • CVPR 2018 • Binh-Son Hua, Minh-Khoi Tran, Sai-Kit Yeung
Deep learning with 3D data such as reconstructed point clouds and CAD models has received great research interests recently.
no code implementations • 8 Sep 2017 • Ramanpreet Singh Pahwa, Minh N. Do, Tian Tsong Ng, Binh-Son Hua
Depth sensing devices have created various new applications in scientific and commercial research with the advent of Microsoft Kinect and PMD (Photon Mixing Device) cameras.
no code implementations • 19 Oct 2016 • Duc Thanh Nguyen, Binh-Son Hua, Lap-Fai Yu, Sai-Kit Yeung
Recent advances of 3D acquisition devices have enabled large-scale acquisition of 3D scene data.
no code implementations • CVPR 2016 • Duc Thanh Nguyen, Binh-Son Hua, Khoi Tran, Quang-Hieu Pham, Sai-Kit Yeung
The proposed method was evaluated on both artificial data and real data obtained from reconstruction of practical scenes.