Ieee Transactions on Pattern Analysis and Machine Intelligence 1 Gradient Response Maps for Real-time Detection of Texture-less Objects

By Stefan Hinterstoißer, Cedric Cagniart, Slobodan Ilic, Peter F. Sturm, Nassir Navab, Pascal Fua, Vincent Lepetit

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

Citations in the corpus

RGB-D datasets using microsoft kinect or similar sensors: a survey (2016, background): “More details about this dataset can be found in [36, 37].”
Cut, Paste and Learn: Surprisingly Easy Synthesis for Instance Detection (2017, background): “These approaches do not work well for objects which are not ‘feature-rich’, where shape-based methods [10, 19, 21] are more successful.”
PoseCNN: A Convolutional Neural Network for 6D Object Pose Estimation in Cluttered Scenes (2017, methodology): “For template-based methods [6], [10], occlusion signiﬁcantly reduces the recognition performance.”
Segmentation-Driven 6D Object Pose Estimation (2018, background): “In the past, textureless objects have often been handled by template-matching [11, 12].”
Global Hypothesis Generation for 6D Object Pose Estimation (2016, methodology): “Template based methods [10, 9, 13] have also been applied to the task of pose estimation.”
Multimodal templates for real-time detection of texture-less objects in heavily cluttered scenes (2011, methodology): “It is based on an efficient representation of templates that capture the different modalities, and we show in many experiments on commodity hardware that our approach significantly outperforms state-of-the-art methods on single modalities.”
6-DoF object pose from semantic keypoints (2017, background): “In recent years, template-based methods have received renewed interest due to the advent of accelerated matching schemes and their ability to detect textureless objects by way of focusing their model description on the object shape [9], [10], [11], [7], [12].”
Implicit 3D Orientation Learning for 6D Object Detection from RGB Images (2018, methodology): “The strategies reach from ignoring one axis of rotation [40,9] over adapting the discretization according to the object [17] to the training of an extra CNN to predict symmetries [28].”
DeepIM: Deep Iterative Matching for 6D Pose Estimation (2018, methodology): “For symmetric objects, we use the closest point distance in computing the average distance for 6D pose evaluation as in Hinterstoisser et al. (2012b): ADD-S = 1 m ∑ x1∈M min x2∈M ‖(Rx1 + t) − (R̂x2 + t̂)‖2.”
Learning descriptors for object recognition and 3D pose estimation (2015, methodology): “Our approach is clearly in the framework of 2D view specific templates [15], which is conceptually sim ple, supported by psychophysical experiments [33], and was successfully applied to various problems and datasets over the last years [22, 19, 13, 11, 8, 28].”
SSD-6D: Making RGB-Based 3D Detection and 6D Pose Estimation Great Again (2017, methodology): “Features can either be an encoding of image properties (color gradients, depth values, normal orientations) [12, 16, 18] or, more recently, the result of learning [4, 29, 5, 6, 17].”
Learning 6D Object Pose Estimation Using 3D Object Coordinates (2014, unknown): “[8, 9, 22].”
Discriminatively Trained Templates for 3D Object Detection: A Real Time Scalable Approach (2013, methodology): “In particular, we build on the work of LINE2D/LINEMOD [5], yet extend it in several directions, applying ideas that have proven useful for category-level recognition to the setting of specific objects: adding a discriminative learning phase for better performance and adding a cascade for better effi”
A codebook-free and annotation-free approach for fine-grained image categorization (2012, methodology): “Using the highly optimized matching technique proposed in [15, 14], we can match each 200 × 300 image with tens of thousands of templates within one second on a standard CPU.”
Model Based Training, Detection and Pose Estimation of Texture-Less 3D Objects in Heavily Cluttered Scenes (2012, methodology): “Since it was already shown in [29] that LINEMOD outperforms DOT [14], HOG [30], TLD [26] and the method of Steger et al.”
BOP: Benchmark for 6D Object Pose Estimation (2018, methodology): “The proposed evaluation methodology formulates the 6D object pose estimation task and deﬁnes a pose-error function which is compared with the commonly used function by Hinterstoisser et al. [13].”
Deep Object Pose Estimation for Semantic Robotic Grasping of Household Objects (2018, methodology): “etection and 6-DoF pose estimation. The problem of object detection and 6-DoF pose estimation from RGB images has been addressed by many researchers using traditional computer vision methods, such as [29,30,31,32]. Recently, several deep learning-based approaches have appeared that operate directly ”
Detecting activities of daily living in first-person camera views (2012, background): “Such a setting may allow one to assume a priori knowledge of the particular objects in the scene, which allows for instance-level visual recognition [16] or RFID-tagged objects [39, 43].”
PVNet: Pixel-Wise Voting Network for 6DoF Pose Estimation (2018, methodology): “Traditional methods mainly rely on template matching techniques [16, 12, 14, 42], which are sensitive to cluttered environments and appearance changes.”