-
Few-Shot Object Detection via Synthetic Features with Optimal Transport
Authors:
Anh-Khoa Nguyen Vu,
Thanh-Toan Do,
Vinh-Tiep Nguyen,
Tam Le,
Minh-Triet Tran,
Tam V. Nguyen
Abstract:
Few-shot object detection aims to simultaneously localize and classify the objects in an image with limited training samples. However, most existing few-shot object detection methods focus on extracting the features of a few samples of novel classes that lack diversity. Hence, they may not be sufficient to capture the data distribution. To address that limitation, in this paper, we propose a novel…
▽ More
Few-shot object detection aims to simultaneously localize and classify the objects in an image with limited training samples. However, most existing few-shot object detection methods focus on extracting the features of a few samples of novel classes that lack diversity. Hence, they may not be sufficient to capture the data distribution. To address that limitation, in this paper, we propose a novel approach in which we train a generator to generate synthetic data for novel classes. Still, directly training a generator on the novel class is not effective due to the lack of novel data. To overcome that issue, we leverage the large-scale dataset of base classes. Our overarching goal is to train a generator that captures the data variations of the base dataset. We then transform the captured variations into novel classes by generating synthetic data with the trained generator. To encourage the generator to capture data variations on base classes, we propose to train the generator with an optimal transport loss that minimizes the optimal transport distance between the distributions of real and synthetic data. Extensive experiments on two benchmark datasets demonstrate that the proposed method outperforms the state of the art. Source code will be available.
△ Less
Submitted 29 August, 2023; v1 submitted 28 August, 2023;
originally announced August 2023.
-
Instance-level Few-shot Learning with Class Hierarchy Mining
Authors:
Anh-Khoa Nguyen Vu,
Thanh-Toan Do,
Nhat-Duy Nguyen,
Vinh-Tiep Nguyen,
Thanh Duc Ngo,
Tam V. Nguyen
Abstract:
Few-shot learning is proposed to tackle the problem of scarce training data in novel classes. However, prior works in instance-level few-shot learning have paid less attention to effectively utilizing the relationship between categories. In this paper, we exploit the hierarchical information to leverage discriminative and relevant features of base classes to effectively classify novel objects. The…
▽ More
Few-shot learning is proposed to tackle the problem of scarce training data in novel classes. However, prior works in instance-level few-shot learning have paid less attention to effectively utilizing the relationship between categories. In this paper, we exploit the hierarchical information to leverage discriminative and relevant features of base classes to effectively classify novel objects. These features are extracted from abundant data of base classes, which could be utilized to reasonably describe classes with scarce data. Specifically, we propose a novel superclass approach that automatically creates a hierarchy considering base and novel classes as fine-grained classes for few-shot instance segmentation (FSIS). Based on the hierarchical information, we design a novel framework called Soft Multiple Superclass (SMS) to extract relevant features or characteristics of classes in the same superclass. A new class assigned to the superclass is easier to classify by leveraging these relevant features. Besides, in order to effectively train the hierarchy-based-detector in FSIS, we apply the label refinement to further describe the associations between fine-grained classes. The extensive experiments demonstrate the effectiveness of our method on FSIS benchmarks. Code is available online.
△ Less
Submitted 14 April, 2023;
originally announced April 2023.
-
The Art of Camouflage: Few-shot Learning for Animal Detection and Segmentation
Authors:
Thanh-Danh Nguyen,
Anh-Khoa Nguyen Vu,
Nhat-Duy Nguyen,
Vinh-Tiep Nguyen,
Thanh Duc Ngo,
Thanh-Toan Do,
Minh-Triet Tran,
Tam V. Nguyen
Abstract:
Camouflaged object detection and segmentation is a new and challenging research topic in computer vision. There is a serious issue of lacking data of camouflaged objects such as camouflaged animals in natural scenes. In this paper, we address the problem of few-shot learning for camouflaged object detection and segmentation. To this end, we first collect a new dataset, CAMO-FS, for the benchmark.…
▽ More
Camouflaged object detection and segmentation is a new and challenging research topic in computer vision. There is a serious issue of lacking data of camouflaged objects such as camouflaged animals in natural scenes. In this paper, we address the problem of few-shot learning for camouflaged object detection and segmentation. To this end, we first collect a new dataset, CAMO-FS, for the benchmark. We then propose a novel method to efficiently detect and segment the camouflaged objects in the images. In particular, we introduce the instance triplet loss and the instance memory storage. The extensive experiments demonstrated that our proposed method achieves state-of-the-art performance on the newly collected dataset.
△ Less
Submitted 21 January, 2024; v1 submitted 14 April, 2023;
originally announced April 2023.
-
Nonlocal Phonon Heat Transport Seen in 1-d Pulses
Authors:
Philip B. Allen,
Nhat Ahn Nghiem Vu
Abstract:
Phonons are the main heat carriers in semiconductor devices. In small devices, heat is not driven by a local temperature gradient, but by local points of heat input and removal. This complicates theoretical modeling. Study of the propagation of vibrational energy from an initial localized pulse provides insight into nonlocal phonon heat transport. We report simulations of pulse propagation in one…
▽ More
Phonons are the main heat carriers in semiconductor devices. In small devices, heat is not driven by a local temperature gradient, but by local points of heat input and removal. This complicates theoretical modeling. Study of the propagation of vibrational energy from an initial localized pulse provides insight into nonlocal phonon heat transport. We report simulations of pulse propagation in one dimension. The 1d case has tricky anomalies, but provides the simplest pictures of the evolution from initially ballistic toward longer time diffusive propagation. Our results show surprising details, such as diverse results from different definitions of atomistic local energy, and failure to exhibit pure diffusion at long times. Boltzmann phonon gas theory, including external energy insertion, is applied to this inherentlytime-dependent and nonlocal problem. The solution, using relaxation time approximation for impurity scattering, does not closely agree with the simulated results.
△ Less
Submitted 26 April, 2022; v1 submitted 1 June, 2021;
originally announced June 2021.
