Papers by Kshitij Agrawal
Cornell University - arXiv, Sep 30, 2019
Autonomous driving relies on deriving understanding of objects and scenes through images. These i... more Autonomous driving relies on deriving understanding of objects and scenes through images. These images are often captured by sensors in the visible spectrum. For improved detection capabilities we propose the use of thermal sensors to augment the vision capabilities of an autonomous vehicle. In this paper, we present our investigations on the fusion of visible and thermal spectrum images using a publicly available dataset, and use it to analyze the performance of object recognition on other known driving datasets. We present an comparison of object detection in night time imagery and qualitatively demonstrate that thermal images significantly improve detection accuracies.
Cornell University - arXiv, Oct 28, 2021
Localization and recognition of less-occurring road objects have been a challenge in autonomous d... more Localization and recognition of less-occurring road objects have been a challenge in autonomous driving applications due to the scarcity of data samples. Few-Shot Object Detection techniques extend the knowledge from existing base object classes to learn novel road objects given few training examples. Popular techniques in FSOD adopt either meta or metric learning techniques which are prone to class confusion and base class forgetting. In this work, we introduce a novel Meta Guided Metric Learner (MGML) to overcome class confusion in FSOD. We reweight the features of the novel classes higher than the base classes through a novel Squeeze and Excite module and encourage the learning of truly discriminative class-specific features by applying an Orthogonality Constraint to the meta learner. Our method outperforms State-of-the-Art (SoTA) approaches in FSOD on the India Driving Dataset (IDD) by upto 11 mAP points while suffering from the least class confusion of 20% given only 10 examples of each novel road object. We further show similar improvements on the few-shot splits of PASCAL VOC dataset where we outperform SoTA approaches by upto 5.8 mAP accross all splits. Preprint. Under review.
A Rare case of giant cell tumor of distal ulna
Final Neural Stem cells Report
Proceedings of the International Conference on Advances in Information Communication Technology & Computing, 2016
This paper presents a new color image encoding and decoding technique using Fractional Fourier Tr... more This paper presents a new color image encoding and decoding technique using Fractional Fourier Transform (FrFT) and Discrete Wavelet Transform (DWT). In proposed work, all three planes of color images are encoded using subbands of DWT and parameters of FrFT. Selection of subband (among the subbands obtained after applying DWT) for applying FrFT and parameters of FrFT are used as security key for the purpose of encoding and decoding of all three color channels. For ensuring the correct decoding of color images, the knowledge of correct selection of subband of DWT for applying FrFT and exact values of FrFT parameters is required. Correct decoding is not possible without the correct knowledge of DWT subband and FrFT parameters values. The proposed technique is also compared with one of the recent existing techniques and experimental results are used to show the effectiveness of the proposed technique.
ArXiv, 2021
Few-shot learning is a problem of high interest in the evolution of deep learning. In this work, ... more Few-shot learning is a problem of high interest in the evolution of deep learning. In this work, we consider the problem of few-shot object detection (FSOD) in a real-world, class-imbalanced scenario. For our experiments, we utilize the India Driving Dataset (IDD), as it includes a class of lessoccurring road objects in the image dataset and hence provides a setup suitable for few-shot learning. We evaluate both metric-learning and meta-learning based FSOD methods, in two experimental settings: (i) representative (same-domain) splits from IDD, that evaluates the ability of a model to learn in the context of road images, and (ii) object classes with less-occurring object samples, similar to the open-set setting in real-world. From our experiments, we demonstrate that the metric-learning method outperforms meta-learning on the novel classes by (i) 11.2 mAP points on the same domain, and (ii) 1.0 mAP point on the open-set. We also show that our extension of object classes in a real-wor...
2021 IEEE/CVF International Conference on Computer Vision Workshops (ICCVW), 2021
Incremental few-shot learning has emerged as a new and challenging area in deep learning, whose o... more Incremental few-shot learning has emerged as a new and challenging area in deep learning, whose objective is to train deep learning models using very few samples of new class data, and none of the old class data. In this work we tackle the problem of batch incremental few-shot road object detection using data from the India Driving Dataset (IDD). Our approach, DualFusion, combines object detectors in a manner that allows us to learn to detect rare objects with very limited data, all without severely degrading the performance of the detector on the abundant classes. In the IDD OpenSet incremental few-shot detection task, we achieve a mAP 50 score of 40.0 on the base classes and an overall mAP 50 score of 38.8, both of which are the highest to date. In the COCO batch incremental few-shot detection task, we achieve a novel AP score of 9.9, surpassing the state-of-the-art novel class performance on the same by over 6.6 times.
Uploads
Papers by Kshitij Agrawal