Parts-based 3D object classification

Huber, Daniel; Kapuria, Anuj; Donamukkala, R.; Hebert, Martial

doi:10.1109/cvpr.2004.1315148

Cited by 111 publications

(77 citation statements)

References 19 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…3D object recognition is still a new research area. Huber et al [10] matched laser rangefinder data to learned object models. Other techniques build 3D object models and match them to still images using local descriptors [11][12][13][14].…”

Section: Introductionmentioning

confidence: 99%

Segmentation and Recognition Using Structure from Motion Point Clouds

Brostow

Shotton

Fauqueur³

et al. 2008

Lecture Notes in Computer Science

784

588

View full text Add to dashboard Cite

Abstract. We propose an algorithm for semantic segmentation based on 3D point clouds derived from ego-motion. We motivate five simple cues designed to model specific patterns of motion and 3D world structure that vary with object category. We introduce features that project the 3D cues back to the 2D image plane while modeling spatial layout and context. A randomized decision forest combines many such features to achieve a coherent 2D segmentation and recognize the object categories present. Our main contribution is to show how semantic segmentation is possible based solely on motion-derived 3D world structure. Our method works well on sparse, noisy point clouds, and unlike existing approaches, does not need appearance-based descriptors. Experiments were performed on a challenging new video database containing sequences filmed from a moving car in daylight and at dusk. The results confirm that indeed, accurate segmentation and recognition are possible using only motion and 3D world structure. Further, we show that the motion-derived information complements an existing state-of-the-art appearance-based method, improving both qualitative and quantitative performance.input video frame reconstructed 3D point cloud automatic segmentation Fig. 1. The proposed algorithm uses 3D point clouds estimated from videos such as the pictured driving sequence (with ground truth inset). Having trained on point clouds from other driving sequences, our new motion and structure features, based purely on the point cloud, perform 11-class semantic segmentation of each test frame. The colors in the ground truth and inferred segmentation indicate category labels.

show abstract

Section: Introductionmentioning

confidence: 99%

Segmentation and Recognition Using Structure from Motion Point Clouds

Brostow

Shotton

Fauqueur³

et al. 2008

Lecture Notes in Computer Science

784

588

View full text Add to dashboard Cite

show abstract

“…The matching score is usually calculated via key-features and spatial descriptors. Such models are reported in [7,8] and show good results using spin-image based descriptors. Additionally, Frome et al [9] introduce 3D and harmonic shape context descriptors, and Mian et al [10] present a matching score which is based on a robust multidimensional table representation of objects.…”

Section: Introductionmentioning

confidence: 99%

Extraction of Objects from Terrestrial Laser Scans by Integrating Geometry Image and Intensity Data with Demonstration on Trees

Barnea

Filin

2012

Remote Sensing

View full text Add to dashboard Cite

Abstract:Terrestrial laser scanning is becoming a standard for 3D modeling of complex scenes. Results of the scan contain detailed geometric information about the scene; however, the lack of semantic details still constitutes a gap in ensuring this data is usable for mapping. This paper proposes a framework for recognition of objects in laser scans; aiming to utilize all the available information, range, intensity and color information integrated into the extraction framework. Instead of using the 3D point cloud, which is complex to process since it lacks an inherent neighborhood structure, we propose a polar representation which facilitates low-level image processing tasks, e.g., segmentation and texture modeling. Using attributes of each segment, a feature space analysis is used to classify segments into objects. This process is followed by a fine-tuning stage based on graph-cut algorithm, which considers the 3D nature of the data. The proposed algorithm is demonstrated on tree extraction and tested on scans containing complex objects in addition to trees. Results show a very high detection level and thereby the feasibility of the proposed framework.

show abstract

“…There are some classification methods that use this information directly [7], [8]. The classification based solely on 3-D information is difficult in the context of the current 3-D reconstruction systems because the information is not accurate enough to allow the reliable extraction of 3-D shapes and surfaces.…”

Section: A Related Workmentioning

confidence: 99%

Multi-feature walking pedestrians detection for driving assistance systems

Bota¹,

Nedesvchi²

2008

IET Intell. Transp. Syst.

View full text Add to dashboard Cite

Abstract-This paper deals with a pedestrian pre-crash sensor based on dense stereovision and motion analysis. We aim to provide means to deploy safety measures before a crash occurs. Depending on the concrete selected actuator, it could be interesting to know the type of object (e. g. vehicle, pedestrian, bicyclist, etc.) involved in the collision. Some actuators need the information about the collision partner type, because they have to be triggered only in specific situations, or in a situation dependent manner. We present a pedestrian detection system based on dense stereo and motion, acting as pre-crash sensor. In order to detect pedestrians, we use simple features such as object size, speed, and also more complex, motion based features. All these features are combined into a flexible, Bayesian framework.

show abstract

Parts-based 3D object classification

Abstract: Abstract

Cited by 111 publications

References 19 publications

Segmentation and Recognition Using Structure from Motion Point Clouds

Segmentation and Recognition Using Structure from Motion Point Clouds

Extraction of Objects from Terrestrial Laser Scans by Integrating Geometry Image and Intensity Data with Demonstration on Trees

Multi-feature walking pedestrians detection for driving assistance systems

Contact Info

Product

Resources

About