paper-with-me

Papers

Feature Intertwiner for Object Detection

2019-03-28 · ICLR 2019 5 · Hongyang Li, Bo Dai, Shaoshuai Shi, Wanli Ouyang, Xiaogang Wang

A well-trained model should classify objects with a unanimous score for every category. This requires the high-level semantic features should be as much alike as possible among samples. To achive this, previous works focus on re-designing the loss or proposing new regularization constraints. In this paper, we provide a new perspective. For each category, it is assumed that there are two feature sets: one with reliable information and the other with less reliable source. We argue that the reliable set could guide the feature learning of the less reliable set during training - in spirit of student mimicking teacher behavior and thus pushing towards a more compact class centroid in the feature space. Such a scheme also benefits the reliable set since samples become closer within the same category - implying that it is easier for the classifier to identify. We refer to this mutual learning process as feature intertwiner and embed it into object detection. It is well-known that objects of low resolution are more difficult to detect due to the loss of detailed information during network forward pass (e.g., RoI operation). We thus regard objects of high resolution as the reliable set and objects of low resolution as the less reliable set. Specifically, an intertwiner is designed to minimize the distribution divergence between two sets. The choice of generating an effective feature representation for the reliable set is further investigated, where we introduce the optimal transport (OT) theory into the framework. Samples in the less reliable set are better aligned with aid of OT metric. Incorporated with such a plug-and-play intertwiner, we achieve an evident improvement over previous state-of-the-arts.

📄 PDF Abstract BibTeX arXiv:1903.11851

Code (2)

hli2020/feature_intertwiner 공식 구현 pytorch
qiu931110/Caffe_OHEM_SoftmaxLoss

Tasks

Objectobject-detectionObject Detection

Methods 이 논문이 사용한 방법론

Average Pooling 설명 없음
Sigmoid Activation 설명 없음
Feature Intertwiner Feature Intertwiner is an object detection module that leverages the features from a more reliable set to help guide the feature learning of another less reliable set. The…
Gradient Clipping One difficulty that arises with optimization of deep neural networks is that large parameter gradients can lead an SGD optimizer to update…
RoIAlign Region of Interest Align, or RoIAlign, is an operation for extracting a small feature map from each RoI in detection and segmentation based tasks. It removes the harsh…
Non Maximum Suppression Non Maximum Suppression is a computer vision method that selects a single entity out of many overlapping entities (for example bounding boxes in object detection). The…
SGD with Momentum 설명 없음
Weight Decay 설명 없음

Similar Papers 제목 키워드 기반

On the Symmetries of Deep Learning Models and their Internal Representations

2022-05-27 · Charles Godfrey, Davis Brown, Tegan Emerson, Henry Kvinge

Symmetry is a fundamental tool in the exploration of a broad range of complex systems. In machine learning symmetry has been explored in both models and data. In this paper we seek to connect the symmetries arising from …

Intertwiners between Induced Representations (with Applications to the Theory of Equivariant Neural Networks)

2018-03-28 · Taco S. Cohen, Mario Geiger, Maurice Weiler

Group equivariant and steerable convolutional neural networks (regular and steerable G-CNNs) have recently emerged as a very effective model class for learning from signal data such as 2D and 3D images, video, and other …

Geometric Deep Learning and Equivariant Neural Networks

2021-05-28 · Jan E. Gerken, Jimmy Aronsson, Oscar Carlsson, Hampus Linander 외

We survey the mathematical foundations of geometric deep learning, focusing on group equivariant and gauge equivariant neural networks. We develop gauge equivariant convolutional neural networks on arbitrary manifolds $\…

Deep Learningobject-detectionObject DetectionSemantic Segmentation

Learning Coordinate-based Convolutional Kernels for Continuous SE(3) Equivariant and Efficient Point Cloud Analysis

2026-03-18 · Jaein Kim, Hee Bin Yoo, Dong-Sig Han, Byoung-Tak Zhang arxiv

A symmetry on rigid motion is one of the salient factors in efficient learning of 3D point cloud problems. Group convolution has been a representative method to extract equivariant features, but its realizations have str…

Semantic Segmentation

Recursive Self-Similarity in Deep Weight Spaces of Neural Architectures: A Fractal and Coarse Geometry Perspective

2025-03-18 · Ambarish Moharil, Indika Kumara, Damian Andrew Tamburri, Majid Mohammadi 외

This paper conceptualizes the Deep Weight Spaces (DWS) of neural architectures as hierarchical, fractal-like, coarse geometric structures observable at discrete integer scales through recursive dilation. We introduce a c…