TS to PNG Converter

How to Convert TS to PNG Online

1. Upload Your TS File: Drag and drop or click "+ Add Files" to load .ts Transport Stream recordings — DVB/ATSC captures, IPTV dumps, HLS segments, or HDV camcorder clips. Batch upload is supported and processing is processed on our servers session.
2. Pick a Frame Selection Mode: Choose Specific Frame and enter a timestamp (e.g., 2.100 for two seconds and 100 milliseconds in) to grab one still, or switch to Multiple Screenshots and set the Framerate (1, 2, 5, 8, 10, 12, 15, 20, 24, 25, 30, or 50 fps — at 1 fps a 30-second clip yields 30 PNGs) to dump a frame sequence.
3. Tune Resolution, Quality, and Compression (Optional): Pick a Resolution Preset (144p through 8K / 4320p) or set custom Width/Height in pixels or percent. Quality Preset defaults to Very High; drop to Medium or Low to shrink output. Adjust the Compression Level slider (1–9, default 6) and Compression Speed (1–9, default 4) for the PNG DEFLATE encoder, optionally reduce the Color Palette (2, 4, 8, 16, 32, 64, 128, 256 colors) and set DPI (72, 96, 150, 200, 300, 400, 600, 1200) for print workflows.
4. Convert and Download: Click "Convert". Single-frame extractions return one PNG; multi-frame jobs return a ZIP of numbered stills. No sign-up, no watermark.

Why Convert TS to PNG?

MPEG Transport Stream (.ts) is the broadcast-and-streaming container defined in ISO/IEC 13818-1 (MPEG-2 Part 1, Systems, ratified July 1995). Each TS packet is exactly 188 bytes with sync markers and error-correction hooks — perfect for unreliable DVB, ATSC, and IPTV channels, but useless when what you actually need is a single still frame from the video stream inside. Converting TS to PNG decodes the H.264, H.265, or MPEG-2 video, extracts frame(s) at your chosen timestamp, and re-encodes them as lossless PNG images.

• Thumbnail TV captures and recordings — Pull a representative still from a DVR / Tvheadend / TVHeadend recording or an HLS .ts segment to use as a video poster, episode thumbnail, or cover art.
• Document broadcast moments — Grab a single PNG from a news or sports TS for a blog post, social share, or legal/archival evidence — PNG's lossless DEFLATE preserves every pixel, unlike JPEG re-encoding.
• Build animation or rotoscoping sequences — Multi-frame export at 24 or 30 fps gives you a numbered PNG sequence (frame_0001.png, frame_0002.png, ...) ready to load into After Effects, Blender, Krita, or DaVinci Resolve as an image sequence.
• Train computer-vision models — Export every Nth frame from surveillance, drone, or dashcam TS captures as PNG for YOLO/Detectron annotation — PNG avoids the JPEG artefacts that confuse object detectors at small object sizes.
• Capture HDV camcorder footage — Older Sony and Canon HDV cameras record 1080i MPEG-2 inside a TS wrapper; grab a still from a wedding or archival tape without re-shooting.
• Diagnose stream corruption — Pulling individual frames at known timestamps from a partially corrupt TS helps locate where DVB packet loss or HLS segment drops actually broke the video.

If you need a smaller still file, try TS to JPG instead — JPEG drops to a fraction of PNG's size for photographic content. Need the whole video repackaged, not stills? See TS to MP4. To shrink the TS itself first, use Compress TS.

TS vs PNG — Format Comparison

Frame Selection Quick Guide

Frequently Asked Questions

What exactly is a .ts file and why doesn't it play in my browser?

.ts is an MPEG Transport Stream — the container the broadcast industry uses for DVB, ATSC, ISDB digital television, and the segments inside an HLS (.m3u8) live stream. It packs video and audio elementary streams into 188-byte packets so receivers can recover from dropped packets on noisy channels. Browsers don't play raw .ts natively because it's a transmission container, not a file-playback one — they expect HLS to be presented through a <video> tag with the manifest, not the standalone segments. Extracting frames as PNG sidesteps all of that.

What's the difference between TS, M2TS, and MTS?

All three are MPEG-2 Transport Stream variants. .ts is the plain 188-byte-packet format from ISO/IEC 13818-1. .m2ts adds a 4-byte timecode header per packet (192-byte packets total) and is the Blu-ray Disc and AVCHD format. .mts is the same M2TS bitstream but with the file extension AVCHD camcorders write straight to SD cards. Our converter accepts plain .ts; for the Blu-ray/AVCHD variants use M2TS to PNG or MTS to PNG.

Can I extract a frame at an exact timestamp with millisecond precision?

Yes — the Time input accepts seconds with three decimals (e.g., 7.250 = 7 seconds 250 ms). The actual returned frame is the nearest displayable frame at or before that timestamp, which is bounded by the source frame rate (at 25 fps the grid is every 40 ms; at 60 fps every ~16.67 ms). Asking for 7.250 in a 25 fps source will return the frame shown at 7.240 — frame-accurate, but limited by what the encoder actually wrote.

Why is my single PNG much larger than I expected?

PNG is lossless, so a 1920×1080 still with photographic content (lots of unique colours, gradients, noise) commonly lands at 2–4 MB even after DEFLATE. Three ways to shrink it: (a) lower the Quality Preset to Medium/Low which scales resolution down before encoding, (b) reduce the Color Palette to 256 or fewer colors so the PNG is written in palette mode (PLTE chunk) instead of truecolor RGB, or (c) convert to JPEG instead via TS to JPG — JPEG typically beats PNG by 5–15x for natural-image content.

How do I pull every frame for use as an image sequence in After Effects or Blender?

Use Multiple Screenshots and set Framerate to match the source (commonly 23.976, 24, 25, 29.97, 30, 50, or 60 fps — our presets cover 1, 2, 5, 8, 10, 12, 15, 20, 24, 25, 30, and 50). The converter returns a ZIP of sequentially numbered PNGs (frame_0001.png, frame_0002.png, …) that After Effects, Premiere, Blender, DaVinci Resolve, and Krita all import as a single image sequence when you tick "Image Sequence" in the import dialog.

Will the PNG preserve interlaced fields from a 1080i broadcast TS?

The converter decodes interlaced TS by deinterlacing first — the PNG you get is a progressive full-frame still, which is what every modern image viewer and editor expects. If you need the raw top/bottom fields as separate images (a niche broadcast-engineering use case), that's not exposed in the web UI; use ffmpeg with -vf separatefields locally.

Can I extract frames from a TS that's still mid-recording or partially corrupt?

Often yes — TS's packet design means a truncated or partially corrupt file usually still has a valid GOP structure up to the point of corruption. Upload it; the converter will extract frames from whatever decodes successfully and skip past unreadable regions. If the entire file is unreadable (failed sync, no PAT/PMT), the conversion will error rather than guess.

Does converting change the resolution or aspect ratio of the original frame?

Default resolution is Original — the PNG comes out at the exact pixel dimensions stored in the TS (typically 1920×1080 for HD broadcasts, 1280×720 for 720p HLS, or 720×480 / 720×576 for SD MPEG-2). If the source uses non-square pixels (anamorphic widescreen, 4:3 SD with PAR 10:11 or 16:11), we correct to square pixels so the PNG displays at the intended aspect on standard image viewers. Pick a Resolution Preset or enter Width/Height to override.

Is anything uploaded to your servers?

Files are processed through your browser session and removed after the session ends. No account, no email, no watermark, and no file-count or duration limits gating the converter.