This paper addresses supervised deep metric learning for open-set image retrieval, focusing on three key aspects: the loss function, mixup regularization, and model initialization. In deep metric learning, optimizing the retrieval evaluation metric, recall@k, via gradient descent is desirable but challenging due to its non-differentiable nature. To overcome this, we propose a differentiable surrogate loss that is computed on large batches, nearly equivalent to the entire training set. This computationally intensive process is made feasible through an implementation that bypasses the GPU memory limitations. Additionally, we introduce an efficient mixup regularization technique that operates on pairwise scalar similarities, effectively increasing the batch size even further. The training process is further enhanced by initializing the vision encoder using foundational models, which are pre-trained on large-scale datasets. Through a systematic study of these components, we demonstrate that their synergy enables large models to nearly solve popular benchmarks.

We present the evaluation methodology, datasets and results of the BOP Challenge 2023, the fifth in a series of public competitions organized to capture the state of the art in model-based 6D object pose estimation from an RGB/RGB-D image and related tasks. Besides the three tasks from 2022 (model-based 2D detection, 2D segmentation, and 6D localization of objects seen during training), the 2023 challenge introduced new variants of these tasks focused on objects unseen during training. In the new tasks, methods were required to learn new objects during a short onboarding stage (max 5 minutes, 1 GPU) from provided 3D object models. The best 2023 method for 6D localization of unseen objects (GenFlow) notably reached the accuracy of the best 2020 method for seen objects (CosyPose), although being noticeably slower. The best 2023 method for seen objects (GPose) achieved a moderate accuracy improvement but a significant 43% run-time improvement compared to the best 2022 counterpart (GDRNPP). Since 2017, the accuracy of 6D localization of seen objects has improved by more than 50% (from 56.9 to 85.6 AR_C). The online evaluation system stays open and is available at: http://bop.felk.cvut.cz/.

For robot manipulation, a complete and accurate object shape is desirable. Here, we present a method that combines visual and haptic reconstruction in a closed-loop pipeline. From an initial viewpoint, the object shape is reconstructed using an implicit surface deep neural network. The location with highest uncertainty is selected for haptic exploration, the object is touched, the new information from touch and a new point cloud from the camera are added, object position is re-estimated and the cycle is repeated. We extend Rustler et al. (2022) by using a new theoretically grounded method to determine the points with highest uncertainty, and we increase the yield of every haptic exploration by adding not only the contact points to the point cloud but also incorporating the empty space established through the robot movement to the object. Additionally, the solution is compact in that the jaws of a closed two-finger gripper are directly used for exploration. The object position is re-estimated after every robot action and multiple objects can be present simultaneously on the table. We achieve a steady improvement with every touch using three different metrics and demonstrate the utility of the better shape reconstruction in grasping experiments on the real robot. On average, grasp success rate increases from 63.3% to 70.4% after a single exploratory touch and to 82.7% after five touches. The collected data and code are publicly available (https://osf.io/j6rkd/, https://github.com/ctu-vras/vishac)

We study the performance of state-of-the-art human keypoint detectors in the context of close proximity human-robot interaction. The detection in this scenario is specific in that only a subset of body parts such as hands and torso are in the field of view. In particular, (i) we survey existing datasets with human pose annotation from the perspective of close proximity images and prepare and make publicly available a new Human in Close Proximity (HiCP) dataset; (ii) we quantitatively and qualitatively compare state-of-the-art human whole-body 2D keypoint detection methods (OpenPose, MMPose, AlphaPose, Detectron2) on this dataset; (iii) since accurate detection of hands and fingers is critical in applications with handovers, we evaluate the performance of the MediaPipe hand detector; (iv) we deploy the algorithms on a humanoid robot with an RGB-D camera on its head and evaluate the performance in 3D human keypoint detection. A motion capture system is used as reference. The best performing whole-body keypoint detectors in close proximity were MMPose and AlphaPose, but both had difficulty with finger detection. Thus, we propose a combination of MMPose or AlphaPose for the body and MediaPipe for the hands in a single framework providing the most accurate and robust detection. We also analyse the failure modes of individual detectors -- for example, to what extent the absence of the head of the person in the image degrades performance. Finally, we demonstrate the framework in a scenario where a humanoid robot interacting with a person uses the detected 3D keypoints for whole-body avoidance maneuvers.

We introduce a loss for metric learning, which is inspired by the Lowe's matching criterion for SIFT. We show that the proposed loss, that maximizes the distance between the closest positive and closest negative example in the batch, is better than complex regularization methods; it works well for both shallow and deep convolution network architectures. Applying the novel loss to the L2Net CNN architecture results in a compact descriptor named HardNet. It has the same dimensionality as SIFT (128) and shows state-of-art performance in wide baseline stereo, patch verification and instance retrieval benchmarks. Papers published at the Neural Information Processing Systems Conference.

We propose a benchmark for 6D pose estimation of a rigid object from a single RGB-D input image. The training data consists of a texture-mapped 3D object model or images of the object in known 6D poses. The benchmark comprises of: i) eight datasets in a unified format that cover different practical scenarios, including two new datasets focusing on varying lighting conditions, ii) an evaluation methodology with a pose-error function that deals with pose ambiguities, iii) a comprehensive evaluation of 15 diverse recent methods that captures the status quo of the field, and iv) an online evaluation system that is open for continuous submission of new results. The evaluation shows that methods based on point-pair features currently perform best, outperforming template matching methods, learning-based methods and methods based on 3D local features. The project website is available at bop.felk.cvut.cz.

We introduce a loss for metric learning, which is inspired by the Lowe's matching criterion for SIFT. We show that the proposed loss, that maximizes the distance between the closest positive and closest negative example in the batch, is better than complex regularization methods; it works well for both shallow and deep convolution network architectures. Applying the novel loss to the L2Net CNN architecture results in a compact descriptor named HardNet. It has the same dimensionality as SIFT (128) and shows state-of-art performance in wide baseline stereo, patch verification and instance retrieval benchmarks.