Top Research Papers on Object Detection

The first systematic study on concealed object detection (COD), which aims to identify objects that are visually embedded in their background, is presented, and rich annotations including object categories, object boundaries, challenging attributes, object-level labels, and instance-level annotations are provided.

A simple but effective framework for COD, termed Search Identification Network (SINet), which outperforms various state-of-the-art object detection baselines on all datasets tested, making it a robust, general framework that can help facilitate future research in COD.

A flexible framework for monocular 3D object detection which explicitly decouples the truncated objects and adaptively combines multiple approaches for object depth estimation, which outperforms the state-of-the-art method.

This paper proposes Pointformer, a Transformer backbone designed for 3D point clouds to learn features effectively, and introduces an efficient coordinate refinement module to shift down-sampled points closer to object centroids, which improves object proposal generation.

This paper proposes a paradigm of lever-aging the contradictory information to enhance the detection ability of both salient object detection and camouflaged object detection, and introduces a "similarity measure" module to explicitly model the contradicting attributes of these two tasks.

This paper tackles the problems of few-shot object detection and few- shot viewpoint estimation, and introduces a simple category-agnostic viewpoint estimation method that outperforms state-of-the-art methods by a large margin on a range of datasets.

This work proposes a novel distillation method for detection tasks based on discriminative instances without considering the positive or negative distinguished by GT, which is called general instance distillation (GID).

This paper proposes an effective adaptive points learning approach to aerial object detection by taking advantage of the adaptive points representation, which is able to capture the geometric information of the arbitrary-oriented instances.

Transformers have become the dominant model in natural language processing, owing to their ability to pretrain on massive amounts of data, then transfer to smaller, more specific tasks via fine-tuning. The Vision Transformer was the first major attempt to apply a pure transformer model directly to images as input, demonstrating that as compared to convolutional networks, transformer-based architectures can achieve competitive results on benchmark classification tasks. However, the computational complexity of the attention operator means that we are limited to low-resolution inputs. For more co...

The extensive experiments show that DiffusionDet achieves favorable performance compared to previous well-established detectors, and possesses an appealing property of flexibility, which enables the dynamic number of boxes and iterative evaluation.

This paper reformulates the knowledge distillation process on localization to present a novel localization distillation method which can efficiently transfer the localization knowledge from the teacher to the student and heuristically introduces the concept of valuable localization region that can aid to selectively distill the semantic and localization knowledge for a certain region.

This work proposes a novel computer vision problem called ORE: Open World Object Detector, which is based on contrastive clustering and energy based unknown identification, and finds that identifying and characterising unknown instances helps to reduce confusion in an incremental object detection setting.

A comprehensive review of data augmentation techniques for object detection is done and problem of class imbalance is outlined with possible solutions and an overview of test time augmentations is presented.

A Double-Head method is proposed, which has a fully connected head focusing on classification and a convolution head for bounding box regression, and it is found that fc-head has more spatial sensitivity than conv-head.

This paper proposes a novel boundary-guided network (BGNet) for camouflaged object detection, which significantly outperforms the existing 18 state-of-the-art methods under four widely-used evaluation metrics.

A Centralized Feature Pyramid (CFP) network for object detection, which is based on a globally explicit centralized feature regulation, which not only has the ability to capture the global long-range dependencies but also efficiently obtain an all-round yet discriminative feature representation.

The goal of COD task is not just to mimic the human visual ability in a single RGB domain, but to go beyond the human biological vision, and the proposed method significantly outperforms other state-of-the-art methods by a large margin.

A multiple center points based learning network (M-CenterNet) is proposed to improve the localization performance of tiny object detection, and experimental results show the significant performance gain over the competitors.

This article extensively reviews this fast-moving research field in the light of technical evolution, spanning over a quarter-century’s time (from the 1990s to 2022).

This work presents a new method that views object detection as a direct set prediction problem, and demonstrates accuracy and run-time performance on par with the well-established and highly-optimized Faster RCNN baseline on the challenging COCO object detection dataset.

This paper systematically study neural network architecture design choices for object detection and proposes a weighted bi-directional feature pyramid network (BiFPN) and a compound scaling method that uniformly scales the resolution, depth, and width for all backbone, feature network, and box/class prediction networks at the same time.

The notion of Prime Samples, those that play a key role in driving the detection performance are proposed, and a simple yet effective sampling and learning strategy called PrIme Sample Attention (PISA) is developed that directs the focus of the training process towards such samples.

A refined YOLOv8 object detection model is proposed, emphasizing motion-specific detections in varied visual contexts, through tailored preprocessing and architectural adjustments, to heighten the model’s sensitivity to object movements.

This work proposes a new instance of joint MOT approach based on Graph Neural Networks (GNNs), which can model relations between variablesized objects in both the spatial and temporal domains, which is essential for learning discriminative features for detection and data association.

A novel Camera-LiDAR Object Candidates (CLOCs) fusion network that provides a low-complexity multi-modal fusion framework that significantly improves the performance of single-modality detectors.

A Depth Calibration and Fusion (DCF) framework that contains two novel components: a learning strategy to calibrate the latent bias in the original depth maps towards boosting the SOD performance and a simple yet effective cross reference module to fuse features from both RGB and depth modalities.

This paper proposes a new method to train object detectors using bounding box annotations for a limited set of object categories, as well as image-caption pairs that cover a larger variety of objects at a significantly lower cost.

In Automatic Driving System (ADS) and Driver Assistance System (DAS), object detection plays a vital part. Nevertheless, existing real-time detection models for tiny vehicle objects have the problems of low precision and poor performance. To solve these issues, we propose a novel real-time object detection model based on You Only Look Once Version 2 (YOLO-v2) deep learning framework for tiny vehicle objects, called Optimized You Only Look Once Version 2 (O-YOLO-v2). In the proposed model, a new structure is introduced to strengthen the feature extraction ability of the network by adding convol...

A union architecture BAD-Net is proposed that connects the dehazing module and detection module in an end-to-end manner and introduces a self-supervised haze robust loss that enables the detection module to deal with different degrees of haze.

We present Pix2Seq, a simple and generic framework for object detection. Unlike existing approaches that explicitly integrate prior knowledge about the task, we cast object detection as a language modeling task conditioned on the observed pixel inputs. Object descriptions (e.g., bounding boxes and class labels) are expressed as sequences of discrete tokens, and we train a neural network to perceive the image and generate the desired sequence. Our approach is based mainly on the intuition that if a neural network knows about where and what the objects are, we just need to teach it how to read t...

A label decoupling framework (LDF) which consists of a label decouple (LD) procedure and a feature interaction network (FIN) which outperforms state-of-the-art approaches on different evaluation metrics.

This work designs a novel integrity learning for SOD, which outperforms the baseline methods in terms of a wide range of metrics and introduces an integrity channel enhancement (ICE) component with the goal of enhancing feature channels that highlight the integral salient objects, while suppressing the other distracting ones.

This review paper focuses on theobject detection algorithms based on deep convolutional neural networks, while the traditional object detection algorithms will be simply introduced as well.

A comprehensive survey of recent advances in visual object detection with deep learning by reviewing a large body of recent related work in literature and covering a variety of factors affecting the detection performance in detail.

The proposed pyramidal feature shrinking network (PFSNet), which aims to aggregate adjacent feature nodes in pairs with layer-by-layer shrinkage, so that the aggregated features fuse effective details and semantics together and discard interference information.

A comprehensive survey of 3D object detection for autonomous driving, encompassing all the main concerns including sensors, datasets, performance metrics and the recent state-of-the-art detection methods, together with their pros and cons.

A single-shot alignment network (S2A-Net) consisting of two modules: a feature alignment module (FAM) and an oriented detection module (ODM) that can achieve the state-of-the-art performance on two commonly used aerial objects’ data sets while keeping high efficiency.

A learning-to-match (LTM) method to break IoU restriction, allowing objects to match anchors in a flexible manner, validating the general applicability of learnable object-feature matching mechanism for visual object detection.

An improved backbone MHSA-Darknet is designed to retain sufficient global context information and extract more differentiated features for object detection via multi-head self-attention and present a simple yet highly effective weighted bi-directional feature pyramid network (BiFPN) for effectively cross-scale feature fusion.

In this paper, the convolution kernels rotate adaptively to extract object features with varying orientations in different images, and an efficient conditional computation mechanism is introduced to accommodate the large orientation variations of objects within an image.