paper-with-me

Papers

BatchFormerV2: Exploring Sample Relationships for Dense Representation Learning

2022-04-04 · Zhi Hou, Baosheng Yu, Chaoyue Wang, Yibing Zhan, DaCheng Tao

Attention mechanisms have been very popular in deep neural networks, where the Transformer architecture has achieved great success in not only natural language processing but also visual recognition applications. Recently, a new Transformer module, applying on batch dimension rather than spatial/channel dimension, i.e., BatchFormer [18], has been introduced to explore sample relationships for overcoming data scarcity challenges. However, it only works with image-level representations for classification. In this paper, we devise a more general batch Transformer module, BatchFormerV2, which further enables exploring sample relationships for dense representation learning. Specifically, when applying the proposed module, it employs a two-stream pipeline during training, i.e., either with or without a BatchFormerV2 module, where the batchformer stream can be removed for testing. Therefore, the proposed method is a plug-and-play module and can be easily integrated into different vision Transformers without any extra inference cost. Without bells and whistles, we show the effectiveness of the proposed method for a variety of popular visual recognition tasks, including image classification and two important dense prediction tasks: object detection and panoptic segmentation. Particularly, BatchFormerV2 consistently improves current DETR-based detection methods (e.g., DETR, Deformable-DETR, Conditional DETR, and SMCA) by over 1.3%. Code will be made publicly available.

📄 PDF Abstract BibTeX arXiv:2204.01254

Code (1)

zhihou7/batchformer pytorch

Tasks

image-classificationImage Classificationobject-detectionObject DetectionPanoptic SegmentationRepresentation Learning

Methods 이 논문이 사용한 방법론

Attention 설명 없음
BatchFormer learn to explore the sample relationships via transformer networks
Linear Layer A Linear Layer is a projection $\mathbf{XW + b}$.
Absolute Position Encodings Absolute Position Encodings are a type of position embeddings for [Transformer-based models] where positional encodings are…
Dropout Dropout is a regularization technique for neural networks that drops a unit (along with connections) at training time with a specified probability $p$ (a common value is…
Softmax The Softmax output function transforms a previous layer's output into a vector of probabilities. It is commonly used for multiclass classification. Given an input vector $x$…
Layer Normalization Unlike batch normalization, Layer Normalization directly estimates the normalization statistics from the summed inputs…
Label Smoothing Label Smoothing is a regularization technique that introduces noise for the labels. This accounts for the fact that datasets may have mistakes in them, so maximizing the…

Similar Papers 제목 키워드 기반

Rethinking Batch Sample Relationships for Data Representation: A Batch-Graph Transformer based Approach

2022-11-19 · Xixi Wang, Bo Jiang, Xiao Wang, Bin Luo

Exploring sample relationships within each mini-batch has shown great potential for learning image representations. Existing works generally adopt the regular Transformer to model the visual content relationships, ignori…

Metric Learning

Self-Supervised Time Series Representation Learning by Inter-Intra Relational Reasoning

2020-11-27 · Haoyi Fan, Fengbin Zhang, Yue Gao

Self-supervised learning achieves superior performance in many domains by extracting useful representations from the unlabeled data. However, most of traditional self-supervised methods mainly focus on exploring the inte…

RelationRelational ReasoningRepresentation LearningSelf-Supervised Learning+3

Team RUC_AIM3 Technical Report at Activitynet 2020 Task 2: Exploring Sequential Events Detection for Dense Video Captioning

2020-06-14 · Yuqing Song, Shi-Zhe Chen, Yida Zhao, Qin Jin

Detecting meaningful events in an untrimmed video is essential for dense video captioning. In this work, we propose a novel and simple model for event sequence generation and explore temporal relationships of the event s…

Dense CaptioningDense Video CaptioningTask 2Video Captioning

Adaptive Mask Sampling and Manifold to Euclidean Subspace Learning with Distance Covariance Representation for Hyperspectral Image Classification

2023-04-07 · IEEE Transactions on Geoscience and Remote Sensing 2023 4 · Mingsong Li, Wei Li, Yikun Liu, Yuwen Huang 외

For the abundant spectral and spatial information recorded in hyperspectral images (HSIs), fully exploring spectral-spatial relationships has attracted widespread attention in hyperspectral image classification (HSIC) co…

Hyperspectral image analysisHyperspectral Image ClassificationHyperspectral Image Segmentationimage-classification+1

SEGIC: Unleashing the Emergent Correspondence for In-Context Segmentation

2023-11-24 · Lingchen Meng, Shiyi Lan, Hengduo Li, Jose M. Alvarez 외

In-context segmentation aims at segmenting novel images using a few labeled example images, termed as "in-context examples", exploring content similarities between examples and the target. The resulting models can be gen…

Meta-LearningOne-Shot SegmentationSegmentationSemantic Segmentation+2