T-LESS: An RGB-D Dataset for 6D Pose Estimation of Texture-Less Objects

By Tomas Hodan, Pavel Haluza, Stepán Obdrzálek, Jiri Matas, Manolis I. A. Lourakis, Xenophon Zabulis. 2017

598 citations recorded by the source corpus.

We introduce T-LESS, a new public dataset for estimating the 6D pose, i.e. translation and rotation, of texture-less rigid objects. The dataset features thirty industry-relevant objects with no significant texture and no discriminative color or reflectance properties. The objects exhibit symmetries and mutual similarities in shape and/or size. Compared to other datasets, a unique property is that some of the objects are parts of others. The dataset includes training and test images that were captured with three synchronized sensors, specifically a structured-light and a time-of-flight RGB-D sensor and a high-resolution RGB camera. There are approximately 39K training and 10K test images from each sensor. Additionally, two types of 3D models are provided for each object, i.e. a manually created CAD model and a semi-automatically reconstructed one. Training images depict individual objects against a black background. Test images originate from twenty test scenes having varying complexity, which increases from simple scenes with several isolated objects to very challenging ones with multiple instances of several objects and with a high amount of clutter and occlusion. The images were captured from a systematically sampled view sphere around the object/scene, and are annotated with accurate ground truth 6D poses of all modeled objects. Initial evaluation results indicate that the state of the art in 6D object pose estimation has ample room for improvement, especially in difficult cases with significant occlusion. The T-LESS dataset is available online at cmp:felk:cvut:cz/t-less.

Subfields: Aerospace Engineering, Computer Vision and Pattern Recognition, Geology

How researchers see this paper

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

In Aerospace Engineering

In Computer Vision and Pattern Recognition

In Geology

Citations in the corpus

Pix3D: Dataset and Methods for Single-Image 3D Shape Modeling (2018, background): “Among those, many object datasets are designed for benchmarking robot manipulation [6, 23, 33, 51].”
Synthetic Data for Deep Learning (2019, background): “[248] provide a real dataset of textureless objects supplemented with 3D models of these objects that provide the 6D ground truth poses.”
Deep Template-based Object Instance Detection (2021, background): “Despite this, efforts have been made to capture such real datasets [3, 13, 16, 33, 7, 34] and to combine them together [17].”
GDRNPP: A Geometry-Guided and Fully Learning-Based Object Pose Estimator (2021, unknown): “Finally, we compare our method withstate-of-the-artmethodsontheBOPbenchmark[34],which containssevencoredatasetsincludingLM-O[86],YCB-V[51], T-LESS[93],TUD-L[94],IC-BIN[95],ITODD[96]and HB[97].”
BB8: A Scalable, Accurate, Robust to Partial Occlusion Method for Predicting the 3D Poses of Challenging Objects without Using Depth (2017, methodology): “We obtain 54% of frames passing the Pose 6D criterion on average on several sequences of the T-LESS dataset, compared to the 67% of the state-of-the-art [10] on the same sequences which uses both color and depth.”
Implicit 3D Orientation Learning for 6D Object Detection from RGB Images (2018, methodology): “4 depicts an exemplary training process for synthetic views of object 5 from T-LESS [15].”
BOP Challenge 2020 on 6D Object Localization (2020, methodology): “object models and images synthesized using the models. While capturing real images of objects under various conditions and annotating the images with 6D object poses requires a signicant human eort [23], the 3D models are either available before the physical objects, which is often the case for manu”
DeepIM: Deep Iterative Matching for 6D Pose Estimation (2018, background): “Furthermore, initial experiments demonstrate that DeepIM is able to accurately match poses for textureless objects (T-LESS [10]) and for unseen objects [28].”
DeceptionNet: Network-Driven Domain Randomization (2019, background): “ signiﬁcant. Apart from the visual gap, supervised approaches also require cumbersome and error-prone human labeling of real training data in form of 2D bounding boxes, segmentation masks or 6D poses [21,12]. For other approaches, such as robotic control learning, solutions must be found by explorat”
CosyPose: Consistent multi-view multi-object 6D pose estimation (2020, methodology): “Our singleview single-object pose estimation method obtains state-of-the-art results on the YCB-Video [18] and T-LESS [19] datasets, achieving a significant 34.”
Few-Shot Object Detection and Viewpoint Estimation for Objects in the Wild (2020, background): “Unlike 6D pose estimation datasets [86], [87], [88] that usually focus on dozens of objects with limited environment variations, Pascal3D+ contains 12 man-made object categories with 2k to 4k images per category,allowing the benchmarking of object pose estimation in the wild.”
BOP: Benchmark for 6D Object Pose Estimation (2018, background): “[16] created a dataset that features industry-relevant objects with symmetries and similarities, and Drost et al.”
Vision-based robotic grasping from object localization, object pose estimation to grasp estimation for parallel grippers: a review (2020, methodology): “Methods of evaluating oriented rectangles Jiang et al. (2011), Vohra et al.”