Data augmentation has shown significant advancements in computer vision to improve model performance over the years, particularly in scenarios with limited and insufficient data. Currently, most studies focus on adjusting the image or its features to expand the size, quality, and variety of samples during training in various tasks including object detection. However, we argue that it is necessary to investigate bounding box transformations as a data augmentation technique rather than image-level transformations, especially in aerial imagery due to potentially inconsistent bounding box annotations. Hence, this letter presents a thorough investigation of bounding box transformation in terms of scaling, rotation, and translation for remote sensing object detection. We call this augmentation strategy NBBOX (Noise Injection into Bounding Box). We conduct extensive experiments on DOTA and DIOR-R, both well-known datasets that include a variety of rotated generic objects in aerial images. Experimental results show that our approach significantly improves remote sensing object detection without whistles and bells and it is more time-efficient than other state-of-the-art augmentation strategies.

Labeling errors in remote sensing (RS) image segmentation datasets often remain implicit and subtle due to ambiguous class boundaries, mixed pixels, shadows, complex terrain features, and subjective annotator bias. Furthermore, the scarcity of annotated RS data due to the high cost of labeling complicates training noise-robust models. While sophisticated mechanisms such as label selection or noise correction might address the issue mentioned above, they tend to increase training time and add implementation complexity. In this paper, we proposeNSegment—a simple yet effective data augmentation solution to mitigate this issue. Unlike traditional methods, it applies elastic transformations only to segmentation labels, varying deformation intensity per sample in each training epoch to address annotation inconsistencies. Experimental results demonstrate that our approach improves the performance of RS image segmentation over various state-of-the-art models.

Mixture-of-Experts (MoE) is a powerful technique for enhancing the performance of neural networks while decoupling computational complexity from the number of parameters. However, despite this, scaling the number of experts requires adding more GPUs. In addition, the load imbalance in token load across experts causes unnecessary computation or straggler problems. We present ES-MoE, a novel method for efficient scaling MoE training. It offloads expert parameters to host memory and leverages pipelined expert processing to overlap GPU-CPU communication with GPU computation. It dynamically balances token loads across GPUs, improving computational efficiency. ES-MoE accelerates MoE training on a limited number of GPUs without degradation in model performance. We validate our approach on GPT-based MoE models, demonstrating 67$\times$ better scalability and up to 17.5$\times$ better throughput over existing frameworks.

The teacher-student (TS) framework, training a (student) network by utilizing an auxiliary superior (teacher) network, has been adopted as a popular training paradigm in many machine learning schemes, since the seminal work—Knowledge distillation (KD) for model compression and transfer learning. Many recent self-supervised learning (SSL) schemes also adopt the TS framework, where teacher networks are maintained as the moving average of student networks, called the momentum networks. This paper presents TSPipe, a pipelined approach to accelerate the training process of any TS frameworks including KD and SSL. Under the observation that the teacher network does not need a backward pass, our main idea is to schedule the computation of the teacher and student network separately, and fully utilize the GPU during training by interleaving the computations of the two networks and relaxing their dependencies. In case the teacher network requires a momentum update, we use delayed parameter updates only on the teacher network to attain high model accuracy. Compared to existing pipeline parallelism schemes, which sacrifice either training throughput or model accuracy, TSPipe provides better performance trade-offs, achieving up to 12.15x higher throughput.

Visual perception of an autonomous vehicle is a crucial component of autonomous driving technologies. While visual perception research has achieved promising performance in recent years, modern methods are mostly trained, applied, and tested on single clean images. Recently, deep learning-based perception methods have addressed multiple degrading effects to reflect real-world bad weather cases, but have achieved only limited success, mainly due to 1) less or no temporal information across adjacent frames and 2) poor correlation between image enhancement and visual perception performance. To solve these issues, in this paper we propose a simple and effective video enhancement network incorporating the perception module, one of the high-level vision tasks, which takes video degraded by adverse weather conditions as an input, and produces an enhanced image and a recognition result as output. Besides, this allows us to leverage temporal information across consecutive frames without flow estimation. We also introduce a new training strategy to robustly guide the high-level task model suitable for both high-quality restoration of images and highly accurate perception. Further, considering sequence-level discrimination, we propose a Sequential Contrastive Loss, called SCL, to maximize apparent discrimination over sequential input. By doing this, our algorithm achieves full interaction showing mutual influence between the enhancement and perception tasks. Further, we introduce a novel low memory network dropping out most of the layer connections of dense blocks to reduce memory usage and computational cost while maintaining high performance. Experiment results demonstrate that the proposed method significantly improves the performance on object detection, distance estimation, and memory usage under adverse weather.