▶️▶️PR動画／▶️Promotional Video

• リモートセンシング、衛星画像処理

• 深層学習を用いた歩行動作解析

• 動的モード分解を介した最適化による動画像復元

• 高ダイナミックレンジ画像の生成

• フラッシュ/ノンフラッシュ画像統合に基づく画像復元

• 固有画像分解(Intrinsic image decomposition)

• 映り込み/雨の軌跡/霧除去

• 車載カメラのための画像復元・補正技術

• スパースFIRフィルタの設計

リモートセンシング、信号処理、画像処理、コンピュテーショナルフォトグラフィ、バイオメカニクスに関する研究に取り組んでいます。

We propose an image restoration technique that uses multiple image integration. The detail of the dark area when acquiring a dark scene is often deteriorated by sensor noise. Simple image integration inherently has the capability of reducing random noises, but it is especially insufficient in scenes that have a dark area. We introduce a novel image integration technique that optimizes the weights for the integration. We find the optimal weight map by solving a proposed convex optimization problem for the weight optimization. Additionally, we apply the proposed weight optimization scheme to a single-image super-resolution problem, where we slightly modify the proposed weight optimization problem to estimate the high-resolution image from a single low-resolution one. We use some of our experimental results to show that the weight optimization significantly improves the denoising and super-resolution performances.

We present a new image integration technique for a flash and long-exposure image pair to capture a dark scene without incurring blurring or noisy artifacts. Most existing methods require well-aligned images for the integration, which is often a burdensome restriction in practical use. We address this issue by locally transferring the colors of the flash images using a small fraction of the corresponding pixels in the long-exposure images. We formulate the image integration as a convex optimization problem with the local linear model. The proposed method makes it possible to integrate the color of the long-exposure image with the detail of the flash image without causing any harmful effects to its contrast, where we do not need perfect alignment between the images by virtue of our new integration principle. We show that our method successfully outperforms the state of the art in the image integration and reference-based color transfer for challenging misaligned data sets. 

物体表面の分光反射率測定は, リモートセンシング, 医療, 食品衛生管理, 物体表面の分光反射率測定は, リモートセンシング, 医療, 食品衛生管理, 植生管理, 生体認証や科学捜査など様々な分野で活用されている. しかしながら, 測定範囲が非常に狭いため被写体の規模が大きくなるにつれて測定効率が低下する問題がある. そこで, 観測画像から画素単位の分光反射率を効率よく推定し保存する技術が求められている. 本研究ではフラッシュ点灯画像を補助情報として, 最適化による分光反射率画像（固有色成分）推定技術を開発し, 色補正問題へ応用することでその有効性を示した.

L0勾配最小化を用いた画像の平滑化手法の研究に取り組んでいます。この平滑化手法を用いることで, より精度の高いトーンマッピングや画像の詳細強調などを実現することができると考えています。

This paper proposes an effective smoothing method based on gradient constraints. Image smoothing based on ℓ0 gradient minimization is useful for some important applications, e.g ., image restoration, intrinsic image decomposition, detail enhancement, and so on. However, undesirable pseudo-edge artifacts often occur in output images. To solve this problem, we introduce novel range constraints in gradient domain. Specifically, the proposed method suppresses these artifacts by introducing appropriate range constraints constructed from a reference image. Experimental results demonstrate the advantages of the proposed method over several conventional methods.

ガラス越しの撮影では手前の風景が撮影した画像中に映り込むことがある. この映り込みの除去は, 撮影者やその周囲の人々のプライバシー保護やロボットビジョンにおける画像認識精度の向上など, 様々な分野において重要な課題である. 本研究では, RGB-D画像を用いた新たな映り込み分離法を提案している.

This paper proposes a novel reflection removal method for RGB-D images that achieve reflection removal and depth map recovery simultaneously. In general, there is a strong structure correlation between an RGB image and a depth map in gradient domain. Based on this fact, we introduce a novel regularization for RGB-D images named the multi-modal structure tensor total variation (MSTV). A proposed minimization problem based on MSTV which is constructed by two minimization problems, reflection removal and depth map recovery, is solved by using alternating direction method of multipliers (ADMM). Experimental results show the effectiveness of our method by applying it to both artificial and real-world images. 

悪天候の場合は雨や雪などによって画面に視野妨害ノイズが映り込み, 人物や車などの画像認識に悪影響を及ぼしてしまう. 画像からこのような不必要な情報を分離し, 明瞭な画像を復元することでパターン認識や異常検出の精度向上が期待できるため, 解決すべき重要な課題である. 本研究では, 露光を変えて撮影した多重画像から雨の軌跡除去と飽和画素復元を同時に達成する新たな画像統合手法を提案する.

屋外で撮影された画像は, 霧や霞, 塵埃などの悪天候によって画像のコントラストが低下するという問題がある. このように悪天候下で撮影された画像から霧を除去し鮮明な画像を取得するための研究が近年活発に行われている. 本研究では近赤外画像と可視画像のペア画像を用いることで, より精度の高い霧除去を目標とする.

This paper proposes a new optimization framework for the joint optimization of sparsity and filter order (JOSFO) for FIR filter design. Since the cost function for JOSFO involves L0 and non-uniform overlapped group  L0 norms, which are not convex, a global optimal solution is difficult to obtain. To find an approximate solution of the non-convex problem, existing approaches repeat the following steps: 1) approximate the cost function; 2) find candidates of zero coefficients by minimizing the cost function; and 3) set them to zero. On the other hand, this paper directly solves the optimization problem, without any approximation to the cost function, by using the alternating direction method of multipliers with the pseudo-proximity operators of  L0 and non-uniform non-overlapped group  L0 norms. Experimental results show that resulting filters designed by the proposed method have sparser coefficients and lower orders, while satisfying filter specifications, such as an error from a desired frequency response.