4.6 A 1/2.3inch 20Mpixel 3-layer stacked CMOS Image Sensor with DRAM

Tsutomu, Haruta; Nakajima, Tasuku; Hashizume, Jiro; Umebayashi, Taku; Takahashi, Hiroshi; Taniguchi, Koki; Kuroda, Masami; Sumihiro, Hiroshi; Enoki, Koji; Yamasaki, Takahiro; Ikezawa, K.; Atsushi, Kitahara; Masao, Zen; Masafumi, Oyama; Koga, Hiroyuki; Tsugawa, Hidenobu; Ogita, Tomoharu; Nagano, Takashi; Takano, Satoshi; Nomoto, Takuya

doi:10.1109/isscc.2017.7870268

Cited by 63 publications

(21 citation statements)

References 3 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…With 3D stacking of top-tier SPADs on advanced digital DSM CMOS, pixel pitches below 8 µm are possible. The recent advent of three tier stacking [ 31 ] allowing the interconnection of back-illuminated sensing layer, processing and memory layers will further enhance the dynamic range, frame rate and power of both SPAD-based and CIS-based oversampled image sensors while facilitating novel ISP approaches such as motion tracking and solid-state optical image stabilization [ 28 ].…”

Section: Discussionmentioning

confidence: 99%

High Dynamic Range Imaging at the Quantum Limit with Single Photon Avalanche Diode-Based Image Sensors

Dutton

Abbas

Gyöngy

et al. 2018

Sensors

View full text Add to dashboard Cite

This paper examines methods to best exploit the High Dynamic Range (HDR) of the single photon avalanche diode (SPAD) in a high fill-factor HDR photon counting pixel that is scalable to megapixel arrays. The proposed method combines multi-exposure HDR with temporal oversampling in-pixel. We present a silicon demonstration IC with 96 × 40 array of 8.25 µm pitch 66% fill-factor SPAD-based pixels achieving >100 dB dynamic range with 3 back-to-back exposures (short, mid, long). Each pixel sums 15 bit-planes or binary field images internally to constitute one frame providing 3.75× data compression, hence the 1k frames per second (FPS) output off-chip represents 45,000 individual field images per second on chip. Two future projections of this work are described: scaling SPAD-based image sensors to HDR 1 MPixel formats and shrinking the pixel pitch to 1–3 µm.

show abstract

Section: Discussionmentioning

confidence: 99%

High Dynamic Range Imaging at the Quantum Limit with Single Photon Avalanche Diode-Based Image Sensors

Dutton

Abbas

Gyöngy

et al. 2018

Sensors

View full text Add to dashboard Cite

show abstract

“…For its sensing element, we emulate the fidelity characteristics of an AR0330 [ 47 ] which is a typical mobile-class image sensor with sufficient number of pixels for providing high-quality images. For its storage element, we emulate the power profile of a 4 Gb LPDDR4 DRAM, which is commonly seen in commercial 3D stacked sensors [ 18 ] for slow-motion video capture. Finally, for its processing element, we emulate the power characteristics of a Myriad2, a vision co-processor found in mobile devices [ 51 , 52 ], capable of neural network processing and feature-based processing, and also Neurostream [ 4 ], another recent candidate architecture for energy-efficient vision processing.…”

Section: Experimental Methodologymentioning

confidence: 99%

“…These imagers still use the traditional CSI interface to communicate with the SoC. For slow-motion capture, Sony [ 18 ] stacks a DRAM beneath the sensor layers. With local memory, the sensor captures and buffers frames at 1000 fps, sending them across the slower camera interface to the host.…”

Section: Background and Related Workmentioning

confidence: 99%

Dynamic Temperature Management of Near-Sensor Processing for Energy-Efficient High-Fidelity Imaging

Kodukula

Katrawala

Jones

et al. 2021

Sensors

View full text Add to dashboard Cite

Vision processing on traditional architectures is inefficient due to energy-expensive off-chip data movement. Many researchers advocate pushing processing close to the sensor to substantially reduce data movement. However, continuous near-sensor processing raises sensor temperature, impairing imaging/vision fidelity. We characterize the thermal implications of using 3D stacked image sensors with near-sensor vision processing units. Our characterization reveals that near-sensor processing reduces system power but degrades image quality. For reasonable image fidelity, the sensor temperature needs to stay below a threshold, situationally determined by application needs. Fortunately, our characterization also identifies opportunities—unique to the needs of near-sensor processing—to regulate temperature based on dynamic visual task requirements and rapidly increase capture quality on demand. Based on our characterization, we propose and investigate two thermal management strategies—stop-capture-go and seasonal migration—for imaging-aware thermal management. For our evaluated tasks, our policies save up to 53% of system power with negligible performance impact and sustained image fidelity.

show abstract

“…Note, with the CMOS-based IS technology, the quality of captured videos using digital cameras has improved significantly. Cameras in new smartphones can record video streams up R f of 960 frames per second (fps) at a resolution of 720p [23]. In addition, there are commercially available high speed cameras with R f of 25.7 kfps and 1 Mfps at resolutions of 1280×800 and 128×32, respectively [24].…”

Section: Introductionmentioning

confidence: 99%

Non-Line-of-Sight MIMO Space-Time Division Multiplexing Visible Light Optical Camera Communications

Hassan

Ghassemlooy

Zvánovec

et al. 2019

J. Lightwave Technol.

View full text Add to dashboard Cite

In this paper, a non-line-of-sight multiple-input multiple-output space and time division multiple access optical camera communications system is proposed for an indoor environment. Mask matching and equal-gain combining (EGC) schemes as well as differential modulation and frame subtraction are used. We propose a unique packet structure to label the transmitters and a new detection method for data extraction from the captured video streams. We outline a comprehensive theoretical model and have developed an experimental testbed to evaluate the performance of the proposed system. The results highlight that zooming and defocusing of the camera does not have a significant impact on the system performance, therefore the aperture can be set to its maximum value. The system performs well over a link span of 10 m with a low transmit power of 12 mW and in the presence of ambient light due to the non-linear conversion of RAW to JPEG. Using mask matching and EGC improves the tolerance of the system to the noise.

show abstract

4.6 A 1/2.3inch 20Mpixel 3-layer stacked CMOS Image Sensor with DRAM

Cited by 63 publications

References 3 publications

High Dynamic Range Imaging at the Quantum Limit with Single Photon Avalanche Diode-Based Image Sensors

High Dynamic Range Imaging at the Quantum Limit with Single Photon Avalanche Diode-Based Image Sensors

Dynamic Temperature Management of Near-Sensor Processing for Energy-Efficient High-Fidelity Imaging

Non-Line-of-Sight MIMO Space-Time Division Multiplexing Visible Light Optical Camera Communications

Contact Info

Product

Resources

About