• 제목/요약/키워드: Fashion images

검색결과 1,314건 처리시간 0.021초

MF sampler: 동영상 기반 패션 검색 모델의 성능 향상을 위한 샘플링 방법 (MF sampler: Sampling method for improving the performance of a video based fashion retrieval model)

  • 백상훈;박종혁
    • 지능정보연구
    • /
    • 제28권4호
    • /
    • pp.329-346
    • /
    • 2022
  • 최근 소셜 미디어의 숏폼(Short form) 동영상(인스타그램, 틱톡, 유튜브) 시장이 점차 증가하면서 인공지능 영역에서는 이를 활용한 연구가 활발히 진행되고 있다. 대표적인 연구분야로 동영상 내의 패션 상품을 탐지하고 상품 이미지를 검색하는 Video to shop 을 들 수 있다. 이와 같은 동영상 기반 인공지능 모델에서는 Convolution 연산을 사용하여 상품의 특징을 추출한다. 하지만 연산 자원의 제한으로 인해, 동영상의 모든 프레임을 사용하여 특징을 추출하는 것은 현실적으로 불가능하다. 이로 인해, 기존 연구에서는 전체 프레임 중 일부만 샘플링해서 사용하거나, 주제의 특성을 활용한 샘플링 방법을 개발하여 이를 통해 위 문제점을 개선하고, 모델의 성능도 향상시켰다. 기존의 Video to shop 연구에서는 프레임을 샘플링 할 때, 무작위로 일부분의 프레임을 샘플링하거나 균등한 간격으로 샘플링 한다. 하지만 이러한 샘플링 방법은 상품이 존재하지 않는 노이즈 프레임을 샘플링 하면서 패션 상품 검색 모델의 성능을 저하시킨다. 이에 본 연구는 노이즈 프레임을 제거하고 검색 모델의 성능을 향상시키는 샘플링 방법 MF(Missing Fashion items on frame) sampler를 제안한다. MF sampler는 키 프레임 메커니즘(Mechanism)을 발전시켜 자원 한계의 문제점을 개선했다. 또한, 노이즈 탐지 모델을 활용한 노이즈 프레임 제거를 통해 검색 모델의 성능을 향상시켰다. 이와 같은 결과는 실험을 통해 확인되었고, Video to shop 패션 상품 검색에 있어 성능 향상과 효과적인 학습이 가능하다는 것을 확인할 수 있었다.

저화질 영상에서 강건한 번호판 추출 방법 (A Robust License Plate Extraction Method for Low Quality Images)

  • 이용우;김현수;강우윤;김경환
    • 전자공학회논문지SC
    • /
    • 제45권2호
    • /
    • pp.8-17
    • /
    • 2008
  • 본 논문에서는 매우 다양한 조건하에서 취득된 영상으로부터 차량 번호판 영역을 추출하는 방법을 제안한다. 제안하는 방법은 색상정보와 에지정보를 보완적으로 사용함으로써 다양한 조명조건은 물론 압축영상에서 흔히 나타나는 영상의 블록화 현상에 강건하게 번호판을 추출한다. 번호판 기울기 각도 보정은 탐색된 영역만을 허프 변환을 통해 각도를 추정, 보정함으로써 연산량을 줄였다. 최종 번호판 후보지 결정은 대상 영역의 가로 대 세로 비율 평가, 수평 스캔하여 영점 교차되는 양상 평가, 레이블링 방법을 이용한 연결 성분 수 측정 등의 방법을 이용하여 결정한다. 다양한 시간대에 수집된 차량 영상을 대상으로 수행된 실험을 통해 제안하는 방법이 강건하게 번호판을 추출 할 수 있음을 확인하였다. 실제 주정차 단속 영상을 대상으로 실험한 결과 94.9%의 번호판 추출 결과를 얻었다.

GLIBP: Gradual Locality Integration of Binary Patterns for Scene Images Retrieval

  • Bougueroua, Salah;Boucheham, Bachir
    • Journal of Information Processing Systems
    • /
    • 제14권2호
    • /
    • pp.469-486
    • /
    • 2018
  • We propose an enhanced version of the local binary pattern (LBP) operator for texture extraction in images in the context of image retrieval. The novelty of our proposal is based on the observation that the LBP exploits only the lowest kind of local information through the global histogram. However, such global Histograms reflect only the statistical distribution of the various LBP codes in the image. The block based LBP, which uses local histograms of the LBP, was one of few tentative to catch higher level textural information. We believe that important local and useful information in between the two levels is just ignored by the two schemas. The newly developed method: gradual locality integration of binary patterns (GLIBP) is a novel attempt to catch as much local information as possible, in a gradual fashion. Indeed, GLIBP aggregates the texture features present in grayscale images extracted by LBP through a complex structure. The used framework is comprised of a multitude of ellipse-shaped regions that are arranged in circular-concentric forms of increasing size. The framework of ellipses is in fact derived from a simple parameterized generator. In addition, the elliptic forms allow targeting texture directionality, which is a very useful property in texture characterization. In addition, the general framework of ellipses allows for taking into account the spatial information (specifically rotation). The effectiveness of GLIBP was investigated on the Corel-1K (Wang) dataset. It was also compared to published works including the very effective DLEP. Results show significant higher or comparable performance of GLIBP with regard to the other methods, which qualifies it as a good tool for scene images retrieval.

HD 영상의 실시간 얼굴 검출을 위한 LBP 연산의 하드웨어 설계 (Hardware Design of LBP Operation for Real-time Face Detection of HD Images)

  • 노현진;김태완;정연모
    • 대한전자공학회논문지SD
    • /
    • 제48권10호
    • /
    • pp.67-71
    • /
    • 2011
  • 디지털 도어락, 디지털 카메라, 비디오 감시 시스템 등에서 사용되는 기존의 얼굴 검출 시스템은 비교적 낮은 해상도의 영상 처리를 사용하고 있으며 이를 위해서 소프트웨어 기반의 구현을 하고 있다. 하지만 이 경우에는 높은 해상도를 위한 얼굴 검출이 어려울 뿐만 아니라 수행해야할 영산 처리 양이 많기 때문에 실시간으로 얼굴을 검출하는데 어려움이 있다. 실시간 임베디드 시스템에서 HD(High Definition) 영상을 위한 효과적인 얼굴 검출을 위해서는 하드웨어적인 접근이 필요하다. 이에 본 논문에서는 얼굴 검출을 위해 사용되는 전처리 과정 중에 하나이며 처리시간이 많이 소요되는 국부 이진 패턴(LBP, Local Binary Pattern) 연산을 하드웨어 구조를 제시하고 설계했다. 그리고 제시한 하드웨어 구조를 FPGA(Field Programmable Gate Array) 칩을 통해서 구현 및 검증을 통해 고해상도 HD급 영상에서 효율적인 얼굴 검출이 가능 한 것을 확인했다.

해체주의를 응용한 메이크업 일러스트레이션 (The Make-up illustrations Based on Deconstruction)

  • 이재열;구자명
    • 한국의류학회지
    • /
    • 제29권3_4호
    • /
    • pp.414-424
    • /
    • 2005
  • This study aims to suggest new directions for the development of make-up design by understanding characteristics of deconstruction make-up and to search for new directions and methods in theme expressions and design conceptions by making illustrations with deconstruction themes. This study examines conceptual characteristics of Derrida's deconstruction, and analyzes the various deconstruction make-up styles through the famous fashion magazines such as Beauty Collection, Collections, Fem, Modain, and Vogue since 1997. The make-up illustration works in this study were made depending on these styles. At first, the nature of deconsouction make-up can be explained as deconstruction of existing methods, deconstruction of the sexes and deconstruction of history or ethnicity. The deconstruction of existing methods can be subdivided into decadence make-up, detester make-up, graphic make-up and collage make-up. Deconstruction of the sexes can be explained androgynous make-up. Deconstruction of history or ethnicity is subdivided into white makeup, ancient Egyptian dark eye make-up and raceless make-up. The deconstruction make-up illustrations present infinite possibilities in expression styles through three dimensional expressions and media mixture using not only drawing materials including watercolor, poster-color, and color pencils but also objets including magazines, coarse fabrics, screentone, beads, suede, sand and so on. The study shows that since deconstruction make-up illustrations are not limited to depiction-centered expressions, they serve creating unique images and future inspirations.

현대 패션에 나타난 금색이미지에 관한 연구 (A Study on the Images of Gold Color in Contemporary Fashion)

  • 김주연;이효진
    • 복식
    • /
    • 제58권3호
    • /
    • pp.123-130
    • /
    • 2008
  • Symbolizing holiness, wealth and power, golden color was most effective to express the absolute power of royal family and aristocratic society as well as the beauty of magnificent and brilliant authority. Such golden color was understood as the artistic color formed in the dictatorial society as well as the society under the rule of absolute monarch. Thus, it has been used in the costume that symbolizes nobleness, holiness, brilliancy, ethnicity and future-oriented property. In this sense, the image of golden color shown in contemporary fashion was analyzed as follows in this research: First, showing the future-oriented image, golden color expresses avant-garde, cyber and cosmic future-oriented image by using diverse new materials which introduce the state-of-the-art technology sense in addition to metallic textile. Second, getting hint from the ethnic costume of many countries around the world such as brilliant embroidery using golden-thread, ornament with exaggerated technique such as gold foil, mark and character in peculiar tint, the ethnic image of golden color expressed the peculiar color or materials of ethnic costume, handicraft details, etc in golden color. Third, since golden color is expressed in ostentatious image to which elegance and luxuriousness that looks like the work of haute-couture costume of royal family or sensuality that voluptuously expresses the beautiful body line of women are added, the viewers are enchanted or attracted.

영화(映畵)의 복식(服飾)과 색채(色彩) 이미지 (A Study on the Costume and Color Image of the Movie)

  • 이예은;조규화
    • 패션비즈니스
    • /
    • 제10권5호
    • /
    • pp.28-44
    • /
    • 2006
  • The purpose of this study is to study the movie costume and color image through the movie "The Aviator". Firstly, looking at the costumes of main characters and their color images expressed in the movie, one extreme of Hughes' image to another was expressed in order to show his extreme inner world by using the light-dark contrast rather than specific colors as the base. The gold color used to express the image of Hepburn represented an image expressing her intellectual elegance. As for her dress silhouette, the taste of intellectual elegance was displayed by a dress whose back was exposed with a grownup-style feminine silhouette or one with drapes and pleats ornaments. To express Gardner's image; that is, her alluring charm and glow, a vivid red color was used. As for her costume, her splendid and fascinating image was expressed by the 40s' rectangular- shouldered tailored suit, exotic peasant blouse and showy accessories. Secondly, the color image expressed in the background in the movie as a special space for Hughes. The red color image of screening room was shown to be as a space for passion and desire and the green color image of men’s room as a space for stability and rest.

조선 민화 나비를 응용한 의상 디자인 연구 (Study on the Fashion Design Applying Pictures of Butterfly in the Korean Folk Paintings)

  • 이종민;이미량
    • 복식문화연구
    • /
    • 제14권5호
    • /
    • pp.828-839
    • /
    • 2006
  • These days, by the accelerated developments of science and industries and the menaces of war and terror, humans have come to have instinct to recur and nostalgia to nature and thereby many art works and designs with natural objectives have begun to be produced. Also the design in 21st century requires creative products based on cultural speciality of the countries. As the examples, we may easily find the images of the countries from visual expressions or products when we get in touch with the products of so called advanced countries in design i. e. U.S.A. Japan, Germany, France, Italy, countries of Scandinavian peninsular. This study, in such stream of the times, was to express our culture's own originality on clothing design with butterfly that is one of natural thing as the material but limited the boundary to realistic butterflies in folk drawings of the times of Chosun. As for the technique of expression, used digital textile printing for best expression of the fine lines, realistic shapes and peculiar colors of the butterflies in Chosun folk drawings and used diverse materials as materials for printing for diversity of the designs. Thereby this thesis is purposed to grope the products with superior competitiveness in the world market by presenting the realistic butterflies in Chosun folk drawings as motives with cultural value native to Korea and applying them to clothing designs.

  • PDF

시판 발열의복의 발열성능 평가 (Evaluation for the Heating Performance of the Heated Clothing on Market)

  • 이현영;정연희
    • 한국의류산업학회지
    • /
    • 제12권6호
    • /
    • pp.843-850
    • /
    • 2010
  • To evaluate the heating performance of commercial heated vests, we investigated the thermal images and the temperature between body and vest for three heated vests. We captured infrared thermography by FT-IR Spectrometer to analyzed the heating temperature of the heating elements taken from the vests, and the maximum heating temperature of the vests was compared with thermal image in the room temperature($18^{\circ}C$). In outdoor experiment($-4.7^{\circ}C$), we measured the inner temperature as well as the thermal image of heated vests. Four healthy men participated in this experiment, and the ANOVA and Duncan test was performed for statistical analysis. As the results, the heating temperature range of the heated vests used in this experiment was $32{\sim}42^{\circ}C$, much lower than the displayed temperature range in their specifications, so the exact specification for heating performance of heated clothing was required. In comparisons of the heating performance among the heated vests, we found out that the insulation of clothing is very important to design the heated clothing, because the inner temperature of the vest had good insulation by itself was higher than that of the vest shown higher temperature over $7^{\circ}$ than another vests at the heating temperature.

외모(外貌)에 있어서 가상(假想) 3D 패션 코디네이션에 대(對)한 질적(質的) 연구(硏究) (A Qualitative Study about Coordination system of 3D Virtual Model)

  • 신효정;김효숙;최창석
    • 패션비즈니스
    • /
    • 제7권2호
    • /
    • pp.143-155
    • /
    • 2003
  • The purpose of this study were to investigate 1) Values placed upon born hereditary appearance and a made up appearance, 2)examined for points of body and clothing and 3)discrepancy in opinions about Coordination system of 3D Virtual Model This study chose qualitative research approach in-depth interviews were from December 12, 2002, to February 20, 2003. The subjects of the study were 12 women aged in twenties 12 women aged over forty. The results of this study were as follows. 1. Women aged in their twenties defined appearance as follows. there are important a made up appearance, Appearance is looking at point of view from body shape to face, makeup, hair style, body image, cosmetic surgery, clothing and is looking at the whole point of view from hair to tiptoe. This seems to include attitude, personality, behavior, images, and feeling. 2. Regarding body image Women aged in their twenties prefer a slender figure. Women aged in their twenties exerts all possible efforts to have an attractive body through dress well. 3. Regarding body image Women aged in their twenties prefer coordinate to system of 3D virtual model.