Chuong Nguyen

Shan Wang, Chuong Nguyen, Jiawei Liu et al.

Reliable segmentation of road lines and markings is critical to autonomous driving. Our work is motivated by the observations that road lines and markings are (1) frequently occluded in the presence of moving vehicles, shadow, and glare and (2) highly structured with low intra-class shape variance and overall high appearance consistency. To solve these issues, we propose a Homography Guided Fusion (HomoFusion) module to exploit temporally-adjacent video frames for complementary cues facilitating the correct classification of the partially occluded road lines or markings. To reduce computational complexity, a novel surface normal estimator is proposed to establish spatial correspondences between the sampled frames, allowing the HomoFusion module to perform a pixel-to-pixel attention mechanism in updating the representation of the occluded road lines or markings. Experiments on ApolloScape, a large-scale lane mark segmentation dataset, and ApolloScape Night with artificial simulated night-time road conditions, demonstrate that our method outperforms other existing SOTA lane mark segmentation models with less than 9\% of their parameters and computational complexity. We show that exploiting available camera intrinsic data and ground plane assumption for cross-frame correspondence can lead to a light-weight network with significantly improved performances in speed and accuracy. We also prove the versatility of our HomoFusion approach by applying it to the problem of water puddle segmentation and achieving SOTA performance.

Shan Wang, Maying Shen, Nadine Chang et al.

Multimodal large language models achieve strong performance across diverse tasks but remain prone to hallucinations, where outputs are not grounded in visual inputs. This issue can be attributed to two main biases: text-visual bias, the overreliance on prompts and prior outputs, and co-occurrence bias, spurious correlations between frequently paired objects. We propose Gradient-based Influence-Aware Constrained Decoding (GACD), an inference-based method, that addresses both biases without auxiliary models, and is readily applicable to existing models without finetuning. The core of our approach is bias estimation, which uses first-order Taylor gradients to understand the contribution of individual tokens-visual features and text tokens-to the current output. Based on this analysis, GACD mitigates hallucinations through two components: (1) suppressing spurious visual features correlated with the output objects, and (2) rebalancing cross-modal contributions by strengthening visual features relative to text. Experiments across multiple benchmarks demonstrate that GACD effectively reduces hallucinations and improves the visual grounding of MLLM outputs.

Hengjia Li, Chuong Nguyen

Image-based 3D reconstruction or 3D photogrammetry of small-scale objects including insects and biological specimens is challenging due to the use of high magnification lens with inherent limited depth of field, and the object's fine structures and complex surface properties. Due to these challenges, traditional 3D reconstruction techniques cannot be applied without suitable image pre-processings. One such preprocessing technique is multifocus stacking that combines a set of partially focused images captured from the same viewing angle to create a single in-focus image. Traditional multifocus image capture uses a camera on a macro rail. Furthermore, the scale and shift are not properly considered by multifocus stacking techniques. As a consequence, the resulting in-focus images contain artifacts that violate perspective image formation. A 3D reconstruction using such images will fail to produce an accurate 3D model of the object. This paper shows how this problem can be solved effectively by a new multifocus stacking procedure which includes a new Fixed-Lens Multifocus Capture and camera calibration for image scale and shift. Initial experimental results are presented to confirm our expectation and show that the camera poses of fixed-lens images are at least 3-times less noisy than those of conventional moving lens images.

Chuong V. Nguyen, David R. Lovell, Matt Adcock et al.

Collections of biological specimens are fundamental to scientific understanding and characterization of natural diversity. This paper presents a system for liberating useful information from physical collections by bringing specimens into the digital domain so they can be more readily shared, analyzed, annotated and compared. It focuses on insects and is strongly motivated by the desire to accelerate and augment current practices in insect taxonomy which predominantly use text, 2D diagrams and images to describe and characterize species. While these traditional kinds of descriptions are informative and useful, they cannot cover insect specimens "from all angles" and precious specimens are still exchanged between researchers and collections for this reason. Furthermore, insects can be complex in structure and pose many challenges to computer vision systems. We present a new prototype for a practical, cost-effective system of off-the-shelf components to acquire natural-colour 3D models of insects from around 3mm to 30mm in length. Colour images are captured from different angles and focal depths using a digital single lens reflex (DSLR) camera rig and two-axis turntable. These 2D images are processed into 3D reconstructions using software based on a visual hull algorithm. The resulting models are compact (around 10 megabytes), afford excellent optical resolution, and can be readily embedded into documents and web pages, as well as viewed on mobile devices. The system is portable, safe, relatively affordable, and complements the sort of volumetric data that can be acquired by computed tomography. This system provides a new way to augment the description and documentation of insect species holotypes, reducing the need to handle or ship specimens. It opens up new opportunities to collect data for research, education, art, entertainment, biodiversity assessment and biosecurity control.

Chuong V Nguyen, Jurgen Fripp, David R Lovell et al.

Thin leaves, fine stems, self-occlusion, non-rigid and slowly changing structures make plants difficult for three-dimensional (3D) scanning and reconstruction -- two critical steps in automated visual phenotyping. Many current solutions such as laser scanning, structured light, and multiview stereo can struggle to acquire usable 3D models because of limitations in scanning resolution and calibration accuracy. In response, we have developed a fast, low-cost, 3D scanning platform to image plants on a rotating stage with two tilting DSLR cameras centred on the plant. This uses new methods of camera calibration and background removal to achieve high-accuracy 3D reconstruction. We assessed the system's accuracy using a 3D visual hull reconstruction algorithm applied on 2 plastic models of dicotyledonous plants, 2 sorghum plants and 2 wheat plants across different sets of tilt angles. Scan times ranged from 3 minutes (to capture 72 images using 2 tilt angles), to 30 minutes (to capture 360 images using 10 tilt angles). The leaf lengths, widths, areas and perimeters of the plastic models were measured manually and compared to measurements from the scanning system: results were within 3-4% of each other. The 3D reconstructions obtained with the scanning system show excellent geometric agreement with all six plant specimens, even plants with thin leaves and fine stems.

Chuong V. Nguyen, Michael Milford, Robert Mahony

Current self-driving car systems operate well in sunny weather but struggle in adverse conditions. One of the most commonly encountered adverse conditions involves water on the road caused by rain, sleet, melting snow or flooding. While some advances have been made in using conventional RGB camera and LIDAR technology for detecting water hazards, other sources of information such as polarization offer a promising and potentially superior approach to this problem in terms of performance and cost. In this paper, we present a novel stereo-polarization system for detecting and tracking water hazards based on polarization and color variation of reflected light, with consideration of the effect of polarized light from sky as function of reflection and azimuth angles. To evaluate this system, we present a new large `water on road' datasets spanning approximately 2 km of driving in various on-road and off-road conditions and demonstrate for the first time reliable water detection and tracking over a wide range of realistic car driving water conditions using polarized vision as the primary sensing modality. Our system successfully detects water hazards up to more than 100m. Finally, we discuss several interesting challenges and propose future research directions for further improving robust autonomous car perception in hazardous wet conditions using polarization sensors.