AIConfPaper
← Search

Juho Kannala

31 accepted papers

2026

Cross-View Splatter: Feed-Forward View Synthesis with Georeferenced Images

Matias Turkulainen, Akshay Krishnan, Filippo Aleotti, Mohamed Sayed, Guillermo Garcia-Hernando, Juho Kannala +3
CVPR 2026

We present Cross-View Splatter, a feed-forward method that predicts pixel-aligned Gaussian splats for outdoor scenes captured at ground level and by satellite. Faithful reconstructions require good camera coverage, but ground imagery is time-consuming and hard to capture at scale for large outdoor s

Cited by 0Source ↗code ↗Scholar ↗
2026

Efficient Reinforcement Learning by Guiding World Models with Non-Curated Data

Yi Zhao, Aidan Scannell, Wenshuai Zhao, Yuxin Hou, Tianyu Cui, Le Chen +4
ICLR 2026poster

Leveraging offline data is a promising way to improve the sample efficiency of online reinforcement learning (RL). This paper expands the pool of usable data for offline-to-online RL by leveraging abundant non-curated data that is reward-free, of mixed quality, and collected across multiple embodime…

Cited by 0Source ↗code ↗Scholar ↗
2026

Predicting Video Slot Attention Queries from Random Slot-Feature Pairs

Rongzhen Zhao, Jian Li, Juho Kannala, Joni Pajarinen
AAAI 2026technical

Unsupervised video Object-Centric Learning (OCL) is promising as it enables object-level scene representation and understanding as we humans do. Mainstream video OCL methods adopt a recurrent architecture: An aggregator aggregates current video frame into object features, termed slots, under some qu

Cited by 0Source ↗PDF ↗Scholar ↗
2025

A Dataset for Semantic Segmentation in the Presence of Unknowns

Zakaria Laskar, Tomas Vojir, Matej Grcic, Iaroslav Melekhov, Shankar Gangisetty, Juho Kannala +3
CVPR 2025poster

Before deployment in the real-world deep neural networks require thorough evaluation of how they handle both knowns, inputs represented in the training data, and unknowns (anomalies). This is especially important for scene understanding tasks with safety critical applications, such as in autonomous…

Cited by 1Source ↗PDF ↗code ↗Scholar ↗
2025

DeSplat: Decomposed Gaussian Splatting for Distractor-Free Rendering

Yihao Wang, Marcus Klasson, Matias Turkulainen, Shuzhe Wang, Juho Kannala, Arno Solin
CVPR 2025poster

Gaussian splatting enables fast novel view synthesis in static 3D environments. However, reconstructing real-world environments remains challenging as distractors or occluders break the multi-view consistency assumption required for accurate 3D reconstruction. Most existing methods rely on external…

Cited by 2Source ↗PDF ↗code ↗Scholar ↗
2025

Multi-Scale Fusion for Object Representation

Rongzhen Zhao, Vivienne Huiling Wang, Juho Kannala, Joni Pajarinen
ICLR 2025poster

Representing images or videos as object-level feature vectors, rather than pixel-level feature maps, facilitates advanced visual tasks. Object-Centric Learning (OCL) primarily achieves this by reconstructing the input under the guidance of Variational Autoencoder (VAE) intermediate representation to…

Cited by 1Source ↗PDF ↗code ↗Scholar ↗
2025

Reloc3r: Large-Scale Training of Relative Camera Pose Regression for Generalizable, Fast, and Accurate Visual Localization

Siyan Dong, Shuzhe Wang, Shaohui Liu, Lulu Cai, Qingnan Fan, Juho Kannala +1
CVPR 2025poster

Visual localization aims to determine the camera pose of a query image relative to a database of posed images. In recent years, deep neural networks that directly regress camera poses have gained popularity due to their fast inference capabilities. However, existing methods struggle to either genera…

Cited by 6Source ↗PDF ↗code ↗Scholar ↗
2024

DGC-GNN: Leveraging Geometry and Color Cues for Visual Descriptor-Free 2D-3D Matching

Shuzhe Wang, Juho Kannala, Daniel Barath
CVPR 2024poster

Matching 2D keypoints in an image to a sparse 3D point cloud of the scene without requiring visual descriptors has garnered increased interest due to its low memory requirements inherent privacy preservation and reduced need for expensive 3D model maintenance compared to visual descriptor-based meth…

Cited by 7Source ↗PDF ↗code ↗Scholar ↗
2024

Differentiable Product Quantization for Memory Efficient Camera Relocalization

Zakaria Laskar*, Iaroslav Melekhov, Assia Benbihi, Shuzhe Wang, Juho Kannala
ECCV 2024poster

"Camera relocalization relies on 3D models of the scene with large memory footprint that is incompatible with the memory budget of several applications. One solution to reduce the scene memory size is map compression by removing certain 3D points and descriptor quantization. This achieves high compr…

Cited by 2Source ↗PDF ↗code ↗Scholar ↗
2024

Efficient NeRF Optimization - Not All Samples Remain Equally Hard

Juuso Korhonen*, Goutham Rangu, Hamed Rezazadegan Tavakoli, Juho Kannala
ECCV 2024poster

"We propose an application of online hard sample mining for efficient training of Neural Radiance Fields (NeRF). NeRF models produce state-of-the-art quality for many 3D reconstruction and rendering tasks but require substantial computational resources. The encoding of the scene information within t…

Cited by 1Source ↗PDF ↗code ↗Scholar ↗
2024

Gaussian Splatting on the Move: Blur and Rolling Shutter Compensation for Natural Camera Motion

Otto Seiskari*, Jerry Ylilammi, Valtteri Kaatrasalo, Pekka Rantalankila, Matias Turkulainen, Juho Kannala +2
ECCV 2024poster

"High-quality scene reconstruction and novel view synthesis based on Gaussian Splatting (3DGS) typically require steady, high-quality photographs, often impractical to capture with handheld cameras. We present a method that adapts to camera motion and allows high-quality scene reconstruction with ha…

Cited by 16Source ↗PDF ↗code ↗Scholar ↗
2024

RP1M: A Large-Scale Motion Dataset for Piano Playing with Bi-Manual Dexterous Robot Hands

Yi Zhao, Le Chen, Jan Schneider, Quankai Gao, Juho Kannala, Bernhard Schölkopf +2
CoRL 2024poster

Endowing robot hands with human-level dexterity is a long-lasting research objective. Bi-manual robot piano playing constitutes a task that combines challenges from dynamic tasks, such as generating fast while precise motions, with slower but contact-rich manipulation problems. Although reinforcemen…

Cited by 2Source ↗Scholar ↗
2024

SuperFusion: Multilevel LiDAR-Camera Fusion for Long-Range HD Map Generation

Hao Dong, Weihao Gu, Xianjing Zhang, Jintao Xu, Rui Ai, Huimin Lu +2
ICRA 2024poster

High-definition (HD) semantic map generation of the environment is an essential component of autonomous driving. Existing methods have achieved good performance in this task by fusing different sensor modalities, such as LiDAR and camera. However, current works are based on raw data or network featu…

Cited by 54Source ↗code ↗Scholar ↗
2023

MixupE: Understanding and improving Mixup from directional derivative perspective

Yingtian Zou, Vikas Verma, Sarthak Mittal, Wai Hoh Tang, Hieu Pham, Juho Kannala +3
UAI 2023poster

Mixup is a popular data augmentation technique for training deep neural networks where additional samples are generated by linearly interpolating pairs of inputs and their labels. This technique is known to improve the generalization performance in many learning paradigms and applications. In this w…

Cited by 11Source ↗PDF ↗code ↗Scholar ↗
2023

Simplified Temporal Consistency Reinforcement Learning

Yi Zhao, Wenshuai Zhao, Rinu Boney, Juho Kannala, Joni Pajarinen
ICML 2023poster

Reinforcement learning (RL) is able to solve complex sequential decision-making tasks but is currently limited by sample efficiency and required computation. To improve sample efficiency, recent work focuses on model-based RL which interleaves model learning with planning. Recent methods further uti…

Cited by 12Source ↗PDF ↗code ↗Scholar ↗
2022

Multiple Offsets Multilateration: A New Paradigm for Sensor Network Calibration with Unsynchronized Reference Nodes

Luca Ferranti, Kalle Åström, Magnus Oskarsson, Jani Boutellier, Juho Kannala
ICASSP 2022accepted

Positioning using wave signal measurements is used in several applications, such as GPS systems, structure from sound and Wifi based positioning. Mathematically, such problems require the computation of the positions of receivers and/or transmitters as well as time offsets if the devices are unsynch…

Cited by 0Source ↗Scholar ↗
2022

Uncertainty-Guided Source-Free Domain Adaptation

Subhankar Roy, Martin Trapp, Andrea Pilzer, Juho Kannala, Nicu Sebe, Elisa Ricci +1
ECCV 2022poster

"Source-free domain adaptation (SFDA) aims to adapt a classifier to an unlabelled target data set by only using a pre-trained source model. However, the absence of the source data and the domain shift makes the predictions on the target data unreliable. We propose quantifying the uncertainty in the…

Cited by 90Source ↗PDF ↗code ↗Scholar ↗
2021

Continual Learning for Image-Based Camera Localization

Shuzhe Wang, Zakaria Laskar, Iaroslav Melekhov, Xiaotian Li, Juho Kannala
ICCV 2021poster

For several emerging technologies such as augmented reality, autonomous driving and robotics, visual localization is a critical component. Directly regressing camera pose/3D scene coordinates from the input image using deep neural networks has shown great potential. However, such methods assume a st…

Cited by 35PDF ↗code ↗Scholar ↗
2021

GraphMix: Improved Training of GNNs for Semi-Supervised Learning

Vikas Verma, Meng Qu, Kenji Kawaguchi, Alex Lamb, Yoshua Bengio, Juho Kannala +1
AAAI 2021technical

We present GraphMix, a regularization method for Graph Neural Network based semi-supervised object classification, whereby we propose to train a fully-connected network jointly with the graph neural network via parameter sharing and interpolation-based regularization. Further, we provide a theoretic…

Cited by 196Source ↗PDF ↗code ↗Scholar ↗
2021

Interpolation-Based Semi-Supervised Learning for Object Detection

Jisoo Jeong, Vikas Verma, Minsung Hyun, Juho Kannala, Nojun Kwak
CVPR 2021poster

Despite the data labeling cost for the object detection tasks being substantially more than that of the classification tasks, semi-supervised learning methods for object detection have not been studied much. In this paper, we propose an Interpolation-based Semi-supervised learning method for object…

Cited by 88PDF ↗code ↗Scholar ↗
2021

Sensor Networks TDOA Self-Calibration: 2D Complexity Analysis and Solutions

Luca Ferranti, Kalle Åström, Magnus Oskarsson, Jani Boutellier, Juho Kannala
ICASSP 2021accepted

Given a network of receivers and transmitters, the process of determining their positions from measured pseudoranges is known as network self-calibration. In this paper we consider 2D networks with synchronized receivers but unsynchronized transmitters and the corresponding calibration techniques, k…

Cited by 0Source ↗Scholar ↗
2020

Hierarchical Scene Coordinate Classification and Regression for Visual Localization

Xiaotian Li, Shuzhe Wang, Yi Zhao, Jakob Verbeek, Juho Kannala
CVPR 2020poster

Visual localization is critical to many applications in computer vision and robotics. To address single-image RGB localization, state-of-the-art feature-based methods match local descriptors between a query image and a pre-built 3D model. Recently, deep neural networks have been exploited to regress…

Cited by 152PDF ↗Scholar ↗
2019

Regularizing Trajectory Optimization with Denoising Autoencoders

Rinu Boney, Norman Di Palo, Mathias Berglund, Alexander Ilin, Juho Kannala, Antti Rasmus +1
NeurIPS 2019poster

Trajectory optimization using a learned model of the environment is one of the core elements of model-based reinforcement learning. This procedure often suffers from exploiting inaccuracies of the learned model. We propose to regularize trajectory optimization by means of a denoising autoencoder tha…

Cited by 16Source ↗PDF ↗Scholar ↗
2018

ADVIO: An Authentic Dataset for Visual-Inertial Odometry

Santiago Cortes, Arno Solin, Esa Rahtu, Juho Kannala
ECCV 2018poster

The lack of realistic and open benchmarking datasets for pedestrian visual-inertial odometry has made it hard to pinpoint differences in published methods. Existing datasets either lack a full six degree-of-freedom ground-truth or are limited to small spaces with optical tracking systems. We take ad…

Cited by 123Source ↗PDF ↗code ↗Scholar ↗
2017

Inertial-based scale estimation for structure from motion on mobile devices

Janne Mustaniemi, Juho Kannala, Simo Särkkä, Jiri Matas, Janne Heikkilä
IROS 2017poster

Structure from motion algorithms have an inherent limitation that the reconstruction can only be determined up to the unknown scale factor. Modern mobile devices are equipped with an inertial measurement unit (IMU), which can be used for estimating the scale of the reconstruction. We propose a metho…

Cited by 26Source ↗Scholar ↗